INQUA-COMMISSION FOR THE STUDY OF THE HOLOCENE Working Group on Data-Handling Methods Newsletter 10, July, 1993 NOTE FROM THE COORDINATOR The Newsletter filled up fast this time, and Dr. Triage is petulant because his column got crowded out again. Eric Grimm's TILIA v. 2.0 is due out in a few months, and I had to promise Triage that he can have several pages to field questions about it! In this issue Glen MacDonald starts the first of several articles on data-handling by CD-ROM. Walter Treloar discusses his work using image analysis to identify pollen. Nancy Marcoux and Pierre Richard describe their technique for handling and displaying information in size-class data of fossil Betula pollen. John Andrews explains an exploratory data analysis program he has created for teasing information out of, for example, long sequences of magnetic-susceptibility readings. Dave Bulman and Peter Kershaw suggest using Chi-Square in attempting to establish modern analogs for fossil pollen in Australia. Minze Stuiver and Paula Reimer have released their new CALIB3, and Paula tells how you can get a free copy by anonymous ftp. The group at the World Data Center-A provide more information about getting data from them, and Eric Grimm issues a call for pollen data sets to add to the North American Pollen Database. Ian Campbell discusses his program CANPLOT, and he also agreed to explain some very interesting things you can do to hand-edit PostScript graphics files. Warren Kovach has issued v. 2.1 of MVSP, and put a shareware version on our INQUA boutique at geology.wisc.edu. Malcolm Clark has added some nice new features to PCSLOT; it also is available from here by ftp. Konrad Gajewski suggests an electronic bulletin board, and I tell how you can join POLPAL-L, a list server for paly- nology. David Green discusses the broad enterprizes of the World Wide Web. I describe SPOON-EZ, a program that connects your computer to an electronic balance and gathers the data for determining the wet and dry weights and volumes of core samples. Owen Davis has sent a concise list of where you can find handy programs for dealing with your data. And last but not least, is the section on e-mail addresses--where you will find that Jim Ritchie, the former newsletter coordinator, is back on the Internet! During the 1993-94 academic year (Sept-May) I am to be on sabbatical to learn as much as I can about the Internet, electronic databases, e-mail, and comput- er uses in paleoecology. I will be in Switzerland and the UK in late August, hope to get to Australia and New Zealand in September, Canada in October and November, Europe again in January and also later in the Spring. If you are doing something you would like to share--or if I could help you with a problem 2of yours--please let me know, and I will try to schedule a visit if I am anywhere near your base of operations. I ask the readers of the Newsletter to send me information on any of the data- handling techniques that you have used which could be helpful to others. Please check your regular and e-mail addresses for accuracy. Send any corrections/suggestions to: Louis J. Maher, Jr. Department of Geology & Geophysics University of Wisconsin 1215 W. Dayton Street Madison, WI 53706 USA Phone: (608) 262-9595 FAX: (608) 262-0693 E-mail: maher@geology.wisc.edu ENVIRONMENTAL DATA BASES ON CD-ROM PART I Glen M. MacDonald Department of Geography McMaster University Hamilton, Ontario Canada L8S 4K1 E-mail: gmmacd@mcmaster.ca Many Quaternary palynologists toil away for endless days counting thousands of pollen grains only to find their prized site reduced to one datum point on a network of sites. They gaze with awe and despair as their data are further amalgamated on maps that look like the stamps of small island nations. Worse still, their beloved data are transformed through one diabolic statistical manipulation or another into an estimate of Thalictrum biomass at 6000 BP or the number of nights in July, 3527 BP when the temperature of freshly formed dew exceeded 15.3755 C. Finally, our exhausted analysts are forced to rest in darkened rooms and ponder enviously the power of Durham, Brown and Boulder. Ah, but take heart! The North American and European Pollen Data Bases are making huge quantities of pollen data available to [*p.1 / p.2*] all. In addition, two of the great technological icons of the late 20th century, the microcomputer and the Compact Disk, can provide even the most humble pollen peeker with untold megabytes of climatic and environmental data with which to compare, manipulate and transfer functionalize all of that pollen data. Oh brave new world! CD-ROM offers an inexpensive and robust way to package very large amounts of data (hundreds of megabytes per disc). It is likely that CD's will replace most tape units for the hard transfer and storage of large data sets. CD's are cheap to produce and can be run on very inexpensive add-on units to micros. At present microcomputer CD's are read only. However, new technologies in the world of audio CD's are being developed for cheap CD recorders. CD-ROM is becoming the standard method for the release of many large data sets. I hope that I can encourage pollen analysts to consider the potential of CD-ROM for providing vast arrays of environmental information for use in conjunction with pollen data. To this end, I will share some of my experiences as a neophyte CD-ROM user. I will not discuss CD technology except to say it is based on laser scanning of a track on the base of the disc and this means "DO NOT SCRATCH THE BASE OF YOUR DISC". Any disturbance of the base can lead to data destruction. In this communication I pass along some observations on using the CD-ROM data set provided by the WORLD WEATHER DISC þ Version 2.1. The WWD is produced by WeatherDisc Associates Inc. and is marketed by the Association of Wildland Fire, P.O. Box 328, Fairfield, Washington, USA, 99012-0328, PH: 509-283-2397, FAX:509-283-2264. The cost is $295 US. Before getting to the WWD it is worth considering what is required to access all of that wonderful data from CD-ROM. In our lab we are using a perfectly ancient DELL 333D (based on a 386 processor). If we can use this museum piece, you can assume that just about any 386- or 486-based micro will run the WWD. Your system should have at least 512 k of memory and run DOS 3.1 or higher. The graphics require EGA display. For those of you in the benighted world of Macintosh, a Mac version was under development when we purchased our DOS version. Our CD-ROM is a Sony CDU6205-10 external unit. You can purchase one of these for $500 US or less. My students inform me that it can be wired to play musical CD's and even display the graphics available from some rock CD's such as the Jimi Hendrix Greatest Hits compilation. I have, thus far, resisted acting on this. You will also require Microsoft Extensions þ 2.0 or later and an appropriate device driver. These may come with your CD-ROM unit. The WWD data are extracted from the archives of the National Climatic Data Center and the National Center for Atmospheric Research in the US. The WWD is mastered to the international ISO 9660 standard and contains approximately 600 megabytes of data. That awesome amount of information provides seventeen individual climate data sets. These data sets range from full station time-series, to mean climate values for airfields, to sea surface temperature time-series. Eleven of the data sets pertain only to the United States. Two data sets are for tropical regions. The remaining four are worldwide. However, read on before you become too enraptured with the endless possibility of pollenclimate transfer functions that this data could allow. Access software is only provided for seven of the data sets. Five of these seven data sets pertain only to the US. The easy to access data sets are: 1) World Monthly Surface Station Climatology: These data are time-series of monthly mean temperature, precipitation and sea-level and/or station pressure for 3265 stations throughout the world. Some records extend back to the 1700's. Many, but not all, contain data up to the mid-1980s. 2) Worldwide Airfield Summaries: This data set includes summaries of monthly and annual temperature, precipitation etc. from 5717 airports throughout the world. The length of record used to calculate means is variable to 1974. 3) US Monthly Normals: These data provide the monthly mean temperature and precipitation normals for 5511 stations in the US. 4) Climatography of the US No. 20: This data set includes normals for a wide variety of climatological measures from 1862 primary, secondary and tertiary stations in the continental US. The normals are calculated for the period 1951-1980. 5) Local Climatological Data: These data are similar to those described above, but contain only 288 primary stations in the US. [*p.2 / p.3*] 6) Climatic Division Data: This data set contains monthly average temperature, precipitation and Palmer drought indices for "climatic regions" (conterminous areas of similar climate) in the US. It is interesting to consider that Pennsylvania is divided into more divisions than California (hmm). The temporal range of these time series is 1895 to the mid 1980's. 7) Daily Weather Observations: These data contain daily values for a number of climatological measures at 205 stations in the US. The time-series generally run from the 1940's to the mid 1980's. In general the access software is very good. For example, in the case of the World Monthly Surface Station data you can select either Imperial or Metric units, then select the continent of interest, then select a specific country and finally select a station. You then choose the parameters you are interested in and view them as color graphs of either monthly values for a specific year or interannual time series of the values for a specific month. The disc access speed is pretty good even on our old beast, and this makes a nice way to scan the climate data from your region of research. Unfortunately, graphic data presentation is not available for all seven data sets for which access software is provided. You can also view the data as tables and export them as ASCII files. Of course, once you have exported the data you can combine it with pollen data to produce transfer functions, re- sponse surfaces or whatever else you can dream up. Power to the masses! The WWD manual is informative and easy to use. In addition, it provides good reference information on the data sets. This includes information on the derivation of the data, appropriate publications for further information and citation, and data structure. The information on data structure is particularly important if you are going to use data sets for which no access software is provided. Despite such a large mass of data and good access software, there are drawbacks with the WWD. Of course, non-US users would like to see more detailed worldwide coverage. For the neophyte, it can be pretty daunting to try to extract information from data sets that are not included in the seven for which WWD provides access software. Many of the individual files in these sets are 60 megabytes or larger! You can forget about calling them up as ordinary Wordperfectþ documents or reading them into many microcomputer stats packs! I hope to discuss this aspect of CD-ROM data in a later communication. Finally, you may find that the records for your favorite station are not complete. For example, for a number of Canadian stations the WWD only provides data up to 1971. Most of these stations are still in operation, and it is a shame not to provide data for the 1970's and 80's. Finally, some of the data sets, such as the Worldwide Airfield Summaries or US Monthly Normals are great candidates for viewing as maps rather than tables and graphs of individual stations. All in all, the WWD is a useful compilation of data. If you do not want to work too hard and are interested in the seven data sets for which access software is provided, it is a good intro to CD-ROM data appropriate for Quaternary palynologists. I suspect that if you are interested in one specific data set and do not mind writing your own access software you could get much of this info through the US government for a nominal charge. In fact, I will look at a large data set coupled with access software that NOAA has released on CD-ROM in my next instalment. Now, where did I leave that Jimi Hendrix CD? AUTOMATION OF POLLEN IDENTIFICATION AND COUNTING USING DIGITAL IMAGE PRO- CESSING TECHNIQUES Walter J. Treloar Postdoctoral Research Fellow Department of Geography Massey University Private Bag Palmerston North New Zealand E-mail : W.J.Treloar@Massey.ac.nz Tel : (06) 356-9099 ext 5423 Fax : (06) 350-5644 Introduction. There are several examples of studies that combine different types of cellular analysis with digital image processing techniques. Systems have been developed for the analysis of genetic material within cells, the sizes of planktonic bacteria or to count blood cells. The research being undertaken at Massey University is to use image processing tech [*p.3 / p.4*] niques to enable the automatic identification and counting of pollen grains. Initially this would have great benefits in areas such as pollen rain, allergy research and apiculture. With further development it is hoped that such a system would be able to identify fossil pollen grains for use within Quaternary Ecology. This would greatly relieve the burden on researchers in the acquisition of raw data and allow greater time to be spent analysing data. Fine resolution sampling could become routine and a greater number of counts per sample would increase the objectivity of the results. The bulk of the initial research into this problem was done in the Departments of Geography and Electronic Engineering at the University of Hull, UK. The research had two foci: the automation of the preparation of pollen samples and pollen identification. Improving the preparation stage was examined first in order to prevent a bottleneck occurring in the automation process. Forster and Flenley (1989) developed complex extraction and purification techniques that could successfully clean out large detrital material and even produced coarse sub-samples of pollen taxa. By largely removing the detrital material before it is analyzed visually means that less time is wasted trying to extract the detrital contaminants at a later stage. In his thesis Mitchel Langford (1989) set down the basis for the current pollen identification research that has recently been incorporated into my own research, first at Hull and now at Massey University. Image Processing System. There are three basic components of an image processing system. These are a video input device, a frame-store (as an add-on within a computer) and a video output monitor. A video input device (such as a video camera or Scanning Electron Microscope) transmits a complete image (frame), line by line, as an analogue video signal where voltage represents brightness. All research undertaken at Hull used a Cambridge Instruments Stereoscan 360 SEM as the input device. A frame-store receives this analogue signal from the input device and converts it to a digital format using a high speed analogue to digital converter. Having been formed by discretising the analogue signal to both brightness and spatial co-ordinates, the digital image is placed line by line in a "rectangular" array of memory in the frame-store where it can be accessed by the computer. It may be considered as a matrix whose column and row indices identify a point in the image, and the corresponding matrix element value identifies the grey-level (or brightness) at that point with black represented by the lowest value and white the highest. The elements of such a digital array are called image elements or pixels. The frame-store used could store an image with dimensions of 256 x 256 pixels and 64 grey-levels. The contents of the frame-store are continually converted back into an analogue signal for display on a video monitor. This aids the user by showing any changes made to the image during processing. Classification Scheme. The current overall classification scheme uses both the preparation and identification stages to automate the classification of pollen. A three stage system is used: 1. The sample preparation stage where the raw pollen sample is treated in various ways to remove as much detrital material as possible. A rough sorting of pollen taxa may also be produced at this level. 2. The intermediate stage level represents the first image processing step of the scheme. The relatively clean pollen sample is mounted under the microscope to be scanned by the computer. Particles are located and then split into sub-samples of pollen and detrital material on the grounds of their general size or shape. Each pollen sub-sample should ideally consist of as few pollen taxa as possible to be analyzed at the final stage. 3. The identification stage of the scheme uses optimal texture features to separate the pollen within each sub-sample. The area of the pollen exine that is used for texture analysis is selected automatically. Two types of texture features are used: the Haralick (Haralick et al., 1973 and Laws Mask (Pietikainen et al., 1983) measures. These produces values that reflect the degree to which an image displays features such as contrast, entropy, waves and ripples at varying scales. The pollen taxa are separated on the grounds of the different texture values produced at this final level. Two different classification techniques are employed so that they can act as counter checks on each other. [*p.4 / p.5*] The first method is a standard statistical classifier, the Fisher Linear Discriminant Function (Hand, 1981), and the second is based upon artificial neural networks (Rumelhart et al., 1986). Results. Initial attempts to classify 12 pollen taxa using only texture features produced very poor results. However the classification scheme outlined above successfully split these 12 pollen taxa into sub-samples consisting of between 1 and 4 taxa each. The sub-samples were correctly identified (using only the statistical method) at the lower level of the scheme at an average of 91% of the time. The majority of the classification rates were 100% but due to a pair of taxa with very poor classification rates the average rate dropped to 91%. The textures of these two taxa were inverses of each other (in the same way that a photograph's inverse is its negative), and the texture measures used were not powerful enough to differentiate between them. The neural network proved to be more efficient than the statistical classifier as 100% classification rates were achieved in all cases except those that included the two taxa described above. Problems Remaining to be Solved. Research here at Massey has now moved away from the use of SEMs for automatic pollen identification. Even though it has been demonstrated that pollen images taken from SEMs are ideal for identification purposes there are draw backs to using them in an automated system. Every image captured from the SEM took at least 45 seconds to construct due to the relatively slow scan rate of the electron beam. Images captured using much faster scan rates were of too poor a quality. This slow scan rate in a final automated system would be unacceptable. Also the cost of a final system, if it incorporated a SEM, would be beyond the financial capabilities of most pollen laboratories. As a result research has now shifted to the use of more conventional light microscopes. An attempt is currently being made to repeat the results achieved on the SEM using a Zeiss Axiophot Photomicroscope. Images may be captured instantaneously from a CCD Video Camera mounted on the microscope. The frame-store used has been upgraded to one with a maximum image dimension of 768 x 512 pixels and 256 grey-levels. Work is also being done to improve and develop the texture discriminators used and thus to drive down the misclassification rate of some taxa. A new and more powerful set of discriminators has now been incorporated and is currently being assessed. With the need to classify a greater number of pollen taxa the present classification scheme may need further refinements. Research is currently under way in the Department of Probability and Statistics at the University of Sheffield to improve the efficiency of the classification scheme. The possible role of neural networks as a solution will also be explored further. The final area that requires work is the automatic control of all the microscope functions including stage movement, lens changing and focusing. Similar systems are already available, for example, in the automatic analysis of cervical smears. Conclusion. There are many areas that still require work before an automated pollen identification system is produced. However, none of them can be said to be unsurmountable. References. Forster, R.M. & Flenley, J.R., 1989. The application of density gradient centrifugation to palynology, School of Geography & Earth Resources, University of Hull, Miscellaneous Series No. 35. Hand, D.J.,1981. Discrimination and Classification, Wiley, Chicester. Haralick, R.M., Shanmagan, K. & Dinstein, I. 1973. Textural features for image classification, IEEE Trans. Systems, Man & Cybernetics, vol. SMC-3, p. 610-621. Langford, M. 1989. Some applications of digital image processing for automation in palynology, Ph.D. thesis, Department of Electronic Engineer- ing, University of Hull. Pietikainen, M., Rosenfeld, A. & Davis, L.S. 1983. Experiments with texture classification using averages of local pattern masks, IEEE Trans. Systems, Man & Cybernetics, vol. SMC-13, p. 421-426. Rumelhart, D.E., McClelland, J.L. et al. 1986. Parallel distributed processing, explorations in the microstructure of cognition, volume 1: foundations, MIT Press, London. [*p.5 / p.6*] A NEW LOOK AT BETULA POLLEN CURVES Nancy Marcoux and Pierre J.H. Richard Laboratoire Jacques-Rousseau Universite de Montreal Montreal, Quebec, Canada E-mail: richard@ere.umontreal.ca There are parts of the world, such as the Gaspe Peninsula (Quebec), northern Scandinavia, and northernmost eastern North America, where Betula pollen grains represent up to 80% of the total terrestrial pollen percentage. In many cases, the species differentiation, commonly based on the size of the pollen grains, becomes interesting for the data interpretation. Usually, the size distribution curves are drawn for each sample. We propose a different graphic representation displaying the percentages curves of the individual diameter values along the entire stratigraphic column. To our knowledge, this display has not been used previously. Lac J'Arrive pollen diagram, which shows a complete postglacial sequence from the northern Gaspe peninsula, will serve as an example. The underlying hypotheses. The idea of measuring the size of Betula pollen grains is not new (Terasmae, 1951; Eneroth, 1951). The goal of this method, sometimes in combination with scuptural and other morphological features (Jentys-Szaferowa, 1928), is to get closer to species identification within the genus in a given regional flora. Various problems are however involved in getting reliable measurements of the equatorial diameter of the grains. Some are related to the chemical treatments of the samples and the nature of the mounting media (Andersen, 1960), and others involve the behaviour of the pollen grains in different sediment types. There are also overlaps in the size of modern Betula grains for different species in a given regional flora (Birks, 1968; Comtois, 1981; Erdtman and Terasmae, 1951; Ives, 1977; Leopold, 1956; Oschurkova, 1959; Prentice, 1981; Richard, 1970; Wenner, 1953). In Quebec, two groups of species are present: dwarf birches (B. glandulosa, B. nana, and B. pumila) and tree birches (B. populifolia, B. borealis, B. papyrifera, B. lenta, and B. alleghaniensis). We do not intend to suggest that all the species may be identified by measurements, since no exhaustive pollen morphological study has been done yet on Betula in Quebec. However, three groups have been recognized according to the size of their pollen: dwarf birches around 20 æm, the first tree species of birches (the B. papyrifera group) around 26 æm, and B. alleghaniensis around 31 æm. The measurements and data preparation. The diameters of 100 triporate, non-folded Betula pollen grains are measured in polar view, at the same magnification routinely used for pollen counting (400X). The slides are mounted in glycerin; consequently, pollen identification and measurements of the Betula grains are done soon after the chemical preparation in order to minimize the effect of the progressive swelling of the grains. Up to now, the measured values were directly rounded to the nearest micrometer unit by the pollen analyst. For higher precision, the results will further be noted as graticule units. The counts of the individual sizes is estimated from the total Betula count. The individual percentages for each size are then calculated (counts per size divided by the total pollen sum) in an Excelþ spreadsheet, .WK1 saved, and transferred to Tiliaùgraph for display. The interpretation for Lac J'Arrive. At lac J'Arrive, the Betula grains are initially a mix of dwarf and tree birches (zone B1), representing long distance transport (see the pollen concentration curve). In zone B2, the tree birches from extra-regional origin are still present but the smallest sizes become regularly represented, probably reflecting the approach of the dwarf birches (B. glandulosa) to the site, the next zone B3 being the principal B. glandulosa period. Zone B4 is a transition zone during which B. glandulosa is gradually replaced by tree birches. In zones B5, greater sizes are well represented, showing the local and regional development of principally B. papyrifera (paper birch). Zone B5b, with the biggest pollen grains diameters, probably corresponds to the development of B. alleghaniensis (yellow birch) in the area. The potential. The routine measurement of thousands of Betula grains in samples from over 150 sites (lakes and bogs) all over Quebec produced results that are internally coherent. The sequence from lac J'Arrive is typical in this regard. In addition, when we find macroremains of Betula, we find the sequential replacement of B. glandulosa by B. papyrifera with a B. alleghaniensis intrusion in some areas of [*p.6 / p.7*] southern Quebec. This original display opens some new perspectives. First, quantitative zoning on Betula sizes can be done to get an objective assessment of the changes. Second, there are possibilities for quantitative assessment of the timing and duration of the replacement of various Betula species based on the size of the grains. In this regard, the identification of the transition zone between the development of successive Betula populations is promising. Third, comparison of the display between diagrams for large regions could reveal geographic patterns in population dynamics within the genus. Even in regions where it is expected that only one Betula specie is involved, the display could be useful in graphing the contribution of each size-unit to the total population of Betula grains. The future. We will probably create a set of Betulasize variables (B15æm, B16æm, ...B40æm) and will enter the counts directly in TILIA. The data will then be stored in the P_Counts table of Paradox, the database manager we currently use (as part of the NAPD_EPD venture). A script in Paradoxþ will handle the appropriate calculations to replace the initial counts by the share of each Betula-size from the total Betula counts. An appropriate designation of the Betula-size will ensure that these counts will be kept outside the total pollen sum. References. Andersen, S.T., 1960. Silicon Oil as a Mounting Medium for Pollen Grains. Danmarks Geologiske Undersogelser, IV Series, 4 (1): 1-24. Andrews, J.T., Mode, W.N., Webber, P.J., Miller, G.H. and Jacobs, J.J., 1980. Report on the distribution of dwarf birches and present pollen rain, Baffin Island, N.W.T., Canada. Arctic, 33: 50-58. Birks, H.J.B., 1968. The identification of Betula nana pollen. New Phytologist, 67 (2): 309-314 + 3 fig., 4 tab. Cherevko, M.V., 1969. On pollen morphology in some species of Betula L. and Alnus Gaertn. Ukrajin, Botanisheskii Zhurnal, 26 (6): 39-45. Claugher, D. and Rowley, J.R., 1987. Betula pollen grain substructure revealed by fast atom etching. Pollen et Spores, 29 (1): 5-20. [*p.7 / p.8*] Clausen, K. E., 1960. A survey of variation in pollen size within individual plants and catkins of three taxa of Betula. Pollen et Spores, 2 (2): 299-304. Comtois, P., 1981. DiamŠtre pollinique de quelques espŠces tourbicoles de Betula. Le Naturaliste canadien, 108 (4): 471-474. Eneroth, O., 1951. Investigations of the possibility of differentiating the pollen of different species of Betula in fossil material. Geologiska Foreningens, Stockholm. Fþrhandlingar., 73 (3): 343-405. Erdtman, G. and Terasm‘, J., 1951. On the pollen morphology of Betula nana. Svensk Botanisk Tidskrift, 45 (2): 358-361. Gaillard, M.-J., 1983. On the occurrence of Betula nana L. pollen grains in the Late-glacial deposits of Lobsigensee (Swiss plateau). Studies in the Late Quaternary of Lobsigensee 2. Revue de Paleobiologie, 2 (2): 181-188. Ives, J.W., 1977. Pollen separation of three North American birches. Arctic and Alpine Research, 9 (1): 73-80. Jentys-Szaferowa, J., 1928. La structure des membranes du pollen de Corylus et Myrica et des espŠces europeennes de Betula et leur determination … l'etat fossile. Bulletin de l'Academie polonaise des Sciences et des Lettres, serie Biologie, Botanique. Johansen, J., 1968. A pollen diagram from the Faroe Islands showing former presence of Betula nana. Annal. Soc. Scient. Faeroensis, 16: 119-128. Leopold, E.B., 1956. Pollen size-frequency in New England species of the genus Betula. Grana, 1 (2): 140-147. Lieux, M.H., 1980. An atlas of pollen of trees, shrubs, and woody vines of Louisiana and other southeastern states. Part II. Platanaceae to Betu- laceae. Pollen et Spores, 22 (2): 191-243. Oschurkova, M.V., 1959. The description of pollen of different species of Betula occuring in the USSR (en russe, resume en anglais). Akademiia nauk SSSR, Moscov; Problemy Botaniki, 4: 68-91. Prentice, I.C., 1981. Quantitative birch (Betula L.) pollen separation by analysis of size frequency data. New Phytologist, 89: 145-157. Richard, P.J.H., 1970. Atlas pollinique des arbres et de quelques arbustes indigŠnes du Qu‚bec. III. Angiospermes (Salicac‚es, Myricac‚es, Juglan- dac‚es, Corylac‚es, Fagac‚es, Ulmac‚es). Le Naturaliste canadien, 97: 97-161 + 29 pl. Terasmae, J., 1951. On the pollen morphology of Betula nana. Svensk Botaniska Tidskrift, 45: 358-361. Usinger, H., 1975. Pollenanalytische und stratigraphische Untersuchungen an zwei Sp„tglazial-Vorkommen in Schleswig-Holstein. Mitteilungen der Arbeitsgemeinschaft Geobotanik in Schleswig-Holstein und Hamburg, 25: 183 p. Vasari, A. and Vasari, Y., 1985. Relic occurence of Betula nana L. in Massif Central, France. Ecologia Mediterranea,11(1): 65-68. Wenner, C.G., 1953. Investigation into the possibilities of distinguishing pollen of the various species of Betula in fossil material. Geologiska Foreningens i Stockholm Forhandlingar, 75 (3): 367-380. CORESEG: A PROGRAM FOR THE DETECTION OF CHANGES (i.e. DEFINITION OF SEGMENTS) IN A DOWN-CORE PROPERTY J. T. Andrews INSTAAR and Department of Geological Sciences, Box 450 University of Colorado, Boulder, CO 80309 E-mail: andrews_jt@cubldr.colorado.edu Introduction. A common and fundamental problem in studies of down-core properties, such as the magnetic susceptibility record of Fig. 1, is to decide where there are "breaks" in the particular variable being considered. Such a change could either be in the average value and/or in the variance about that value. Webster examined this problem and proposed a solution to it (Webster, 1973; Webster, 1980); the general problem is also discussed by Davis (1986). My program CORESEG was initially developed without knowledge of the earlier papers by Webster (1973, 1980), but it turns out to adopt a similar approach with some twists (Andrews and Stravers, in press). It was originally written for a TRS80 computer using standard BASIC; the present version is in use on a MAC SE/30 and LCIII (QuickBASIC). [*p.8 / p.9*] Details of the program. Appendix I leads you through the questions posed as the program is executed. A major difference between this program and that of Webster (1980) is that CORESEG uses a lot of code that was written for Exploratory Data Analysis (EDA) and published in 1981 (Velleman and Hoaglin, 1981). Thus CORESEG uses the median as the measure of central tendency and derives an estimate of the standard deviation from the spread between the upper (L3) and lower hinges (L2) of the sample (i.e. the 25 and 75 percentiles). The standard deviation for a series is estimated from: s = (L3-L2) * 1.349 The program computes the median (md) and the estimated standard deviation (s) for a forward (f) and backward (b) window of length l and then derives a generalized distance measure D2 which is computed as: D2 = (Mdf-Mdb)2/(s2f + s2b) ... Fig. 1. Volume magnetic susceptibility data (MS, x 10-5 SI) from marine piston core HU90-023-001, Frobisher Bay, NWT, Canada (Andrews and Stravers, in press). The next two columns show the strength of D2 with window lengths 10 and 20, respectively, and the far column shows the location of possible segment boundaries. Thus one of the decisions that is made during the program's running is the length of the forward and backward window. Of course there is no reason to limit the analysis to a single length, l, and I have found it informative to run a variety of window lengths and then compare the results (Fig. 1). If a Mac is being used, the output of D2 values can be exported to the "clipboard" and then imported to a variety of graphical/statistical packages for plotting and further analysis. The D2 values are partly related to the length of the window, but a critical question is: which peaks in D2 represent significant changes in the variable and denote the start or end of a new segment? To assist in obtaining a "feel" for the answer to this question CORESEG proceeds to compute a random series with the same mean and standard deviation as the original time-series and then computes a D2 series for this random series. This series can then be plotted as a probability plot (Fig. 2) and a conservative (i.e. high) value for D2 chosen such that the original D2 parameter might be only expected to exceed a particular value 1 in 100 trials. Note that the quasi-random generator available on most personal computers does not generate a new random time-series on each reiteration unless a new seed is specified ... Fig. 2. Probability plots of D2 values for window lingths 10 and 20 (see Fig. 1) and for a random time-series with the same mean and standard deviation (HU#001 RND D2). On the right are box-plots of the D2 parameters. The program is written so that one transform (a log transform) is an option; frequently magnetic susceptibility is plotted on a log scale. In addition, the first difference operator can be called (see Appendix 1) and used to investigate the importance of a trend in the data (Fig. 3). The various 8measures of D2 can be [*p.9 / p.10*] saved to disk as can the random series and its D2 series. ... Figure 3. Magnetic susceptibility (same units as Fig. 1) for HU90-023-022 from Breevort Basin (Andrews and Stravers, in press) and the D2 plot for these data after the first difference operator had been used; compare the upper graph of D2 (on the 1st difference) versus the comparison between it (solid line) and D2 with a window of 10 (lower figure, dashed line). In the original version a variety of output was routed to the printer, but in this code modified for the Mac there is no printer output, instead files are written to disk. The data results, shown as three different "Results" (see Appendix 1), include output on the median, the hinges, maximum values etc. In the Mac version these need to be written down as the program proceeds. For those with an IBM then the output can be sent to the printer with the "LPRINT" command. As with any analytical tool this program is not a panacea for all concerns involved with describing and segmenting a core. However, in my experience it does give a number of useful insights into down-core changes, and it focuses attention on specific levels within the cores. References. Andrews, J. T., and Stravers, J. A., in press, Magnetic susceptibility of late Quaternary marine sediments, Frobisher Bay, N.W.T.: An indicator of ice sheet/ocean interactions: Quaternary Science Reviews, v. p. Davis, J. C., 1986, Statistics and data analysis in Geology: New York, John Wiley & Sons, 646 p. Velleman, P. F., and Hoaglin, D. C., 1981, Applications, Basics, and Computing of Exploratory Data Analysis: Boston, Duxbury, 354 pp p. Webster, R., 1973, Automatic soil-boundary location from transect data: Journal International Association Mathematical Geology v. 5, p. 27-37. Webster, R., 1980, DIVIDE: A FORTRAN IV Program for segmenting multivariate one-dimensional spatial series:Computers and Geosciences v. 6, p. 61-68. APPENDIX Line No. Question > Comments or Action 475 Max. length of record? > Number of items in input file 2440 Data input-AGAIN OR STOP > To start, type AGAIN 4540 File name? > File name (1 item/line in ASCII) 2565 Use the first diff. operator Y/N? > N=as is, Y=removes trend 2590 1. Results on input data > Shows median, hinges etc. 2640 Data need a transform Y/N? > N=as is, Y=log transform 3940 Proceed with analysis of core seg. Y/N? 2730 Print out all results,Y=0,N=1 > Type 1 2740 Treat standardized data Y/N? > Ignore, type N 3150 Length of window? > Length must be > 3 4920 Storing r2 values as file/r2 4922 File name for d2=? > Type File name 3640 2. Results on test statistic > Shows median, hinges, etc. of d2 3655 Continue Y/N? > Type Y; N takes you to 2440 4000 Results 3 random= > Results shown 4001 File name random nos? > Type File name 4005 Continue? > Type Y [*p.10 / p.11*] 4920 Storing re values as file/r2 4922 File name for d2=? > Type File name for random d2 ON THE USE OF THE CHI-SQUARE STATISTIC FOR THE DETECTION OF MODERN ANALOGUES FOR FOSSIL POLLEN SPECTRA Dave Bulman and Peter Kershaw Centre for Palynology and Palaeoecology Dept. of Geography and Environmental Science Monash University Clayton, Vic. 3168, Australia E-mail: dbulman@arts.cc.monash.edu.au geg625n@vaxc.cc.monash.edu.au We have been investigating the potential for using a recent pollen data set, composed of pre-European pollen spectra from some 71 pollen diagrams, for determination of past vegetation and climate analogues from earlier Holocene spectra within these diagrams in southeastern Australia. The sites from which the pollen diagrams have been constructed are scattered throughout the shaded area on Fig. 1; a preliminary examination of the recent spectra will appear shortly in the Review of Palaeobotany and Palynology. From an examination of different dissimilarity measures, Overpeck et al. (1985) concluded that squared-chord distance is the most efficient for the determination of modern analogues from fossil spectra. This has subsequently been used in a number of studies and appears to be a component of what is developing as established methodology for analogue detection. However, we feel that this measure is not appropriate for our purposes for two major reasons. ... Fig. 1. Index Map 1. It puts proportionately greater emphasis on minor than major taxa in the records. This would be very acceptable in southeastern Australia, where much of the vegetation and consequently the derived pollen spectra are dominated by few ecologically wide ranging taxa, particularly Eucalyptus and Poaceae, if it was not for the relatively low level of reliability in representation of minor taxa. In a number of diagrams, especially those produced a number of years ago before the establishment of good modern pollen reference collections, a number of minor taxa were not recognised. Even when recognised, the generally low pollen counts, resulting at least partly from low pollen influx values, have prohibited determination of accurate percentage levels. It is therefore necessary to place emphasis on the major taxa. 2. The establishment of suitable "cut-off" levels of dissimilarity using squared-chord distance for each fossil spectrum is determined from the range of variation within modern pollen spectra from each identified present day vegetation type. In the case of southeastern Australia, with only 71 sites, it is not possible to get anything like the ten (as suggested by Overpeck et al. 1985) modern sites within any particular vegetation type so that some other cut-off method for analogue identification is required. Instead of squared-chord distance, we have opted to use Chi-square, as proposed by Mosimann (1965) for use with percentage pollen data, which is also considered to be a fairly good measure of dissimilarity. This measure treats all taxon percentages in relation to their true numerical value, and cut-off levels can be determined from the usual statistical confidence levels for the appropriate degrees of freedom. Before we take our analyses too far, it would be valuable to receive comments on the use of Chi-square under these circum- stances and suggestions on the existence of any, more appropriate, techniques. Please send responses and requests for further details of the project to Dave Bulman at the address above. References. Mosimann, J. E. 1965. Statistical Methods for the Pollen Analyst. In B. Kummel and D. Raup (Eds.). Handbook of Paleontological Techniques. Freeman and Co., San Francisco, pp. 636-673. [*p.11 / p.12*] Overpeck, J.T., Webb III, T., and Prentice, I.C. 1985. Quantitative interpretation of fossil pollen spectra: dissimilarity coefficients and the method of modern analogs. Quaternary Research 23, 87-108. CALIBRATE YOUR 14C DATES WITH THE LATEST VERSION OF CALIB Louis J. Maher Minze Stuiver and Paula Reimer recently distributed an up-dated version of CALIB, their calibration program for 14C dates. CALIB (Stuiver & Reimer, 1986) was originally released in 1986 and went through revisions 2.0 and 2.1 in 1987. CALIB 3.0 (Stuiver & Reimer, 1993) was released this Spring, with version 3.0.3 current as this newsletter is being assembled. If you work with carbon dates, you will definitely want to get this program. In addition to a number of refinements, it includes a new calibration data set covering nearly 22,000 cal yr (÷18,400 14C yr). Several laboratories have contributed to a six-year calibration effort to produce the new data and to correct the old. CALIB on disk (IBM & clones) is available free of charge from: Quaternary Isotope Lab, c/o Ms. Paula Reimer or Dr. Minze Stuiver, University of Washing- ton, Seattle, WA 98195; (206) 543-6327, FAX: (206) 543-3836; E-mail: minze@u.washington.edu or pjreimer@u.washington.edu. But if you are on the Internet you can get the latest version at any time by anonymous ftp. The address is ftp.u.washington.edu (or 140.142.56.2). CALIB.EXE and related files, including the manual in ASCII and also Microsoft WORD format are compressed in a self-extracting file named CALIB3.EXE in the sub-directory /public/calib. [There is also a subdirectory pub, but you want public. Ed.] You need to ftp ftp.u.washington.edu [or ftp 140.142.56.2]. Logon as anonymous and use your e-mail address as the password. Set the file transfer type to binary by typing the word binary. Change to CALIB3.EXE's sub- directory: cd /public/calib and get the file by typing: get calib3.exe . When you have the binary file on your IBM-compatible computer's drive, you can extract the 11 files (totaling 622,718 bytes; 874,071 bytes with CALIB3.EXE) by typing the word calib3 . If you do not want to extract the files immediately, but rather wish to test whether the package arrived safely, type: calib3 /t and the following information should appear on the screen: PKSFX (R) FAST! Self Extract Utility Version 1.1 03-15-90 Copr. 1989-1990 PKWARE Inc. All Rights Reserved. PKSFX/h for help PKSFX Reg. U.S. Pat. and Tm. Off. Searching EXE: C:/CALIB3.EXE Testing: CALIB.EXE OK Testing: UWTEN93.14C OK Testing: INTCAL93.14C OK Testing: MARINE93.14C OK Testing: COURB.FON OK Testing: MSHERC.COM OK Testing: RESIDUAL.14C OK Testing: MANUAL.DOC OK Testing: SAMPLE1A.A00 OK Testing: CALS&R.DAT OK Testing: MANUAL.TXT OK If you wish to extract just the ASCII manual, type: calib3 man*.txt and browse through it at your leisure. Paula Reimer asked me to say that unfortunately CALIB for the Macintosh is not available at this time. They had unexpected difficulties with the Mac version and may have to rewrite the code in c. It will not be available until further notice. References. Stuiver, M., and Reimer, P. J., 1986. A computer program for radiocarbon age calibration, Radiocarbon 28: 1022-1030. Stuiver, M., and Reimer, P. J., 1993. Extended 14C database and revised CALIB radiocarbon calibration program, Radiocarbon 35: 215-230. [*p.12 / p.13*] NEW BOOKSHELF 7 H.J.B. Birks E-mail: birks@cc.uib.no The following recently published books may be of interest to readers of this Newsletter. Bell, M. & Walker, M.J.C. 1992 Late Quaternary environmental change. Physical and human perspectives. Longman, Harlow. 273 pp. Bradley, R.S. & Jones, P.D. (Eds.) 1992 Climate since A.D. 1500. Routledge, London & New York. 679 pp. Brooker, P.I. 1991 A geostatistical primer. World Scientific, Singapore. 95 pp. Chambers, F.M. (Ed.) 1993 Climatic change and human impact on the landscape. Chapman and Hall, London. 303 pp. Coker, P.D. 1992 Inside the grey box. Choosing, building and upgrading a PC. I/O Press, Leyburn. 258 pp. Paperback. Everitt, B.S. 1993 Cluster Analysis (Third Edition). Edward Arnold, London. 170 pp. Francis, B. et al. 1993 The GLIM System 4. The Statistical System for Generalised Linear Interactive Modelling. Clarendon Press, Oxford. 821 pp. Paperback. Frenzel, B. (Ed.) 1992 European climate reconstructed from documentary data: methods and results. European Science Foundation Project European Palaeoclimate and Man 2, 265 pp. Strasbourg. Paperback. Frenzel, B. (Ed.) 1992 Evaluation of land surfaces cleared from forests by prehistoric man in Early Neolithic times and the time of the migrating Germanic tribes. European Science Foundation Project European Palaeoclimate and Man 3, 225 pp. Strasbourg. Paperback. Glen-Lewin, D.C., Peet, R.K. & Veblen, T.T. (Eds.) 1992 Plant succession. Theory and Prediction. Chapman and Hall, London. 352 pp. Goodess, C.M., Palutikof, J.P. & Davies, T.D. 1992 The nature and causes of climate change. Assessing the long-term future. Belhaven Press, London. 248 pp. Goudie, A. 1992 Environmental Change. Clarendon Press, Oxford. 329 pp. Paperback. Hall, P. 1992 The bootstrap and Edgeworth expansion. Springer-Verlag, New York. 352 pp. Jones, R.L. & Keen, D.H. 1993 Pleistocene environments in the British Isles. Chapman and Hall, London. 346 pp. Paperback. Newton, H.J. 1988 TIMESLAB: a time series analysis laboratory. Wadsworth & Brooks/Cole, Pacific Grove, California. 623 pp. Paperback with diskette (up-date diskette available from the author). Solomon, A.M. & Shugart, H.H. 1993 Vegetation dynamics and global change. Chapman and Hall, London and IIASA. 338 pp. Paperback. Tufte, E.R. 1990 Envisioning information. Graphics Press, Cheshire, Connecticut. 126 pp. Watson, D.F. 1992 Contouring. A Guide to the Analysis and Display of Spatial Data. Pergamon Press, Oxford. 321 pp. With diskette. Williams, M.A.J., Dunkerley, D.L., DeDekker, P., Kershaw, A.P. & Stokes, T. 1993 Quaternary Environments. Edward Arnold, London. 329 pp. Paperback. Yarnal, B. 1993 Synoptic climatology in environmental analysis. Belhaven Press, London. 195 pp. Please take a moment to check that your mail address is accurate. If your e- mail address is listed at the end of this Newsletter, please check it too. E-mail the Coordinator any changes that should be made. [*p.13 / p.14*] NORTH AMERICAN POLLEN DATABASE CALL FOR DATA Eric C. Grimm Illinois State Museum Research and Collections Center 1920 South 10 1/2 Street Springfield, IL 62703, USA E-mail: grimm@denr1.igis.uiuc.edu The North American Pollen Database is established at the Illinois State Museum and with funding from the NOAA Paleoclimatology Program. Data are entered into the database at the Illinois State Museum and periodically delivered to the National Geophysical Data Center (NGDC) in Boulder, Colorado. The data may be acquired from NGDC for free via FTP. Most of the data now in the database were acquired from COHMAP, although some new data has been added. The COHMAP data is largely pre-1980. We are now actively soliciting and adding new data. We, of course, prefer the data in electronic form (almost any format is acceptable, although Tilia files are great). Data can be sent on diskette or by email. We also welcome data on paper. We understand that not all workers have the time or resources to computerize their old data. WORLD DATA CENTER-A FOR PALEOCLIMATOLOGY AT THE NOAA PALEOCLIMATOLOGY PROGRAM: July 1993 Update Robert S. Webb rsw@mail.ngdc.noaa.gov Jonathan T. Overpeck jto@mail.ngdc.noaa.gov David M. Anderson dma@mail.ngdc.noaa.gov Bruce A. Bauer bab@mail.ngdc.noaa.gov NOAA National Geophysical Data Center Paleoclimatology Program/World Data Center-A for Paleoclimatology 325 Broadway, E/GC, Boulder, CO, 80303, USA. The World Data Center-A (WDC-A), located at the U.S. National Oceanic and Atmospheric Administration (NOAA) National Geophysical Data Center (NGDC) Paleoclimatology Program, continues to expand its role as coordination center for paleoenvironmental data management activities needed by the international research community. The program is working closely with the International Geosphere-Biosphere Program/Past Global Changes (IGBP/PAGES) and will help coordinate the PAGES Core Project Data Management Workshop "Global Paleoenvironmental Data", Bern, Switzerland, August, 1993. The goals of the workshop will be to establish protocols for data sharing and exchange, coordinate database efforts, and broaden participation throughout the international community. The WDC-A continues to solicit contributions of all paleoenvironmental datasets for archiving and distribution. In the last three months the data center has received a number of datasets including: * Dr. W.R. Peltier contributed the global database of sea level histories from 20,000 yr B.P. to present from the Tushingham and Peltier (1992) ICE-3G study. * Dr. W.R. Peltier also contributed a 1øx1ø global database of icesheet reconstructions at 1000-year intervals from 21,000 yr B.P. to present that is being used as boundary conditions in the IGBP/PAGES Paleoclimate Modeling Intercomparison Project (PMIP). * Dr. J. Imbrie contributed the SPECMAP 3 archive for distribution synchronous with its publication in Nature (Imbrie et al., 1993). This is a timely release of data and is a good example for all in the scientific community! * Drs. G. Shen, J. Cole, and D. Lea and contributed additional geochemical data from annually banded corals in the Galapagos Islands. * Dr. K. Gajewski contributed a dataset of climate reconstructions from laminated sediments lakes in the northeastern and midwestern U.S.A. * The dendrological community continue to support the International Tree-Ring Data Bank managed by the Paleoclimatology Program. To date this year a total of 14 researchers have contributed tree-ring data. The establishment of the Anonymous FTP/Internet server to obtain all archived paleoclimate datasets free of charge has been a remarkable success. In the first six months of operation, 375 users from 15 countries downloaded over 750 megabytes of data (two-thirds of all data distribution). One of the most popular datasets has been the North American Pollen Database (NAPD). The NAPD can be accessed and downloaded following these basic instructions: [*p.14 / p.15*] To LOGON: FTP NGDC1.NGDC.NOAA.GOV -or- FTP 192.149.148.121 Enter anonymous as your login name, and your e-mail address as password. You should now be connected and see an FTP> prompt. (NOTE: If your version of FTP does not support multiple line messages enter " - " before your email address/password. Your system may return slightly different messages and prompts) SAMPLE FTP SESSION FTP> cd paleo (change to /paleo directory) FTP> cd pollen (change to /pollen directory) FTP> ls (lists 2 documentation files: [pollen.doc & pollen.readme] and 2 subdirecto8ries [asciifiles & zipfiles] on the screen. Within the subdirectory asciifiles are two subdirectories: modern and fossil; the modern subdirectory contains "m50" files of modern surface samples contain- ing counts for the 50 most abundant pollen types; the fossil subdirectory contains "t50" files of fossil stratigraphies containing counts for the 50 most abundant pollen types; and "p15" files of fossil stratigraphies containing percentage data for the 15 most abundant pollen taxa at each site. Within the subdirectory zipfiles are 6 DOS-PKZIP files: fos_p15.zip, fos_t50.zip, mod_all.zip, mod_byst.zip, napd_doc.zip, and napd_tbl.zip.) FTP> get pollen.readme (copy readme file to your computer) FTP> mget *.* (copy all files within the current subdirectory to your computer) FTP> cd (change back to the "root" directory) FTP> cd pub (change to the public directory) FTP> put mydata (send to NGDC your file "mydata") FTP> quit (end FTP session) The WDC-A for Paleoclimatology remains committed to try to accept contributions of digital paleoenvironmental data in any logical file format from Macintosh, DOS, or UNIX machines. Data on magnetic media or over FTP/INTERNET are acceptable. Preferred formats for contributing data are: 1) standard ITRDB ("Tucson format") for tree ring data, 2) Tiliaùgraph for pollen data, 3) CLIMAP structured ASCII files for deep sea fauna, 4) tab or space delimited ASCII, 5) commercial spreadsheets (e.g., LOTUS, EXCEL, with complete documentation). For information and advice on how to contribute your data contact Bruce Bauer (phone: 303 497 6280; fax: 303 497 6513; e-mail and mailing address above). For information on the program, ordering data on tape or diskettes, or to be added to the mailing list contact Mrs. Mildred England (phone: 303 497-6227; mailing address above; e-mail: mke@mail.ngdc.noaa.gov). KEEP THE HEART OF THE NAPD ON A SINGLE HD FLOPPY Louis Maher The National Geophysical Data Center's World Data Center-A has made available ASCII-text files of 293 fossil pollen sites in the North American Pollen Database (Webb et al. 1993; see also Webb et al., this newsletter). The files contain data from either the top 15 taxa or the top 50 taxa--as percentage data or as the actual counts. If you take the raw counts of the top 50 taxa, you are getting the heart of the NAPD with which you can start doing science almost immediately. I took a copy of fos_t50.zip from WDC-A by anonymous ftp. The original uncompressed *.T50 files total 4,018,548 bytes. They start out with 11 lines of information identifying site and author. They indicate the number of taxa and levels in the file, list the taxa by short as well as full name, and then provide the counts. As far as I am aware, none of the common computer programs can read and interpret these files directly--although they can be read by any word-processor. I considered changing my POLFILE.EXE so it could be used to read and edit the *.50 files, but it seemed more efficient to change the whole lot over to my *.RAW format. The Wisconsin *.RAW format allows them to be imported directly into Eric C. Grimm's TILIA. They can be read directly by my POLFILE.EXE and converted to *.DAT files for use with PLOTSITE, SLOTSEE, or [*p.15 / p.16*] SLOTDEEP, or they can be output in a format that can be read by Warren L. Kovach's MultiVariate Statistical Package: MVSP. I wrote a short program called NAPD2RAW.BAS which would read a *.T50 file and convert it to a *.RAW file; then adapted it so that--given a list of the *.T50 filenames--it would automatically read and convert them all. In the process I discarded the taxa abbreviations and kept the whole name. In a few cases when the names were long, I shorted them. For example: "Pinus subgen. Haploxylon" became "Pinus.Hap"; "Asteraceae subfam. Asteroideae undiff." became "Asteraceae.undif."; "Chenopodiaceae/Amaranthaceae" became "Cheno/Am"; and "other trees + shrubs" became "Oth.trees/shrubs". My *.RAW files put documentation after the numerical data, so the NAPD header was attached at the end. As an example, the header for DEVILSWI.T50 which was appended to DEVILSWI.RAW is reproduced below (The lines indicated by "*" were added during the conversion): *# Source: # NAPD ASCII Format # 07 Jan 93 # Top 50 Pollen Types # # Site name: Devils Lake [Wisconsin] # Place: USA:Wisconsin # Latitude: 43.25.00 # Longitude: -89.44.00 # Elevation(m): 294 # Contact person: Maher, L.J., Jr. # Chron name: COHMAP chron# 2 *# Latitude Degrees: 43.41667 *# Longitude Degrees: -89.73333 *# AGE ESTIMATES ARE LINEAR APPROXIMATIONS *# USING SELECTED DATES; *# CONSULT ORIGINAL DATES AND STRATIGRAPHY *# FOR CRITICAL WORK. *# Converted to WISCONSIN *.RAW format with *# NAPD2RAW: 02-08-1993 The *.T50 files separate degree, minute, and second coordinates with periods. I use decimal degrees in some of my plotting, so I added that form as well. All but 4 files (Cupola, Jackson, Weber-C1, Weber-S1) have age estimates for each level. I have reproduced these as "Est.Age(BP)", and also added a category "Est.Age(BP)Decades)". When *.DAT files are made with POLFILE.EXE, substituting "Est.Age(BP)Decades)" for "Depth(cm)" will allow PLOTSITE.EXE's depth ticks to indicate time in 1000's of years BP. The conversion resulted in a reduced storage requirement. The original uncompressed *.T50 files total 4 mb whereas the *.RAW files occupy 1.5 mb. I zipped the 293 files into a 644,039-byte self-extracting file called T50RAW.EXE which is now available for anonymous ftp in /pub/inqua (ftp geology.wisc.edu or ftp 144.92.137.14). Be sure to type binary before issuing the get command; you need to download an exact copy for binary files to work. I have also added to /pub/inqua a copy of POLFILE.EXE in the self-extracting file called POLFIL15.EXE. This will allow you to edit the T50 *.RAW files. You can keep T50RAW.EXE on a HD floppy disk and unzip the files you want whenever you need them. One of the reasons I favor keeping archived files as self-extracting *.EXE files is that they can be expanded by anyone. If they are *.zip files, the latest PKUNZIP will unzip the work of earlier versions, but the converse is not true. (PKZIP þ and PKUNZIP þ are the products of PKWARE Inc., 9025 N. Deerwood Dr., Brown Deer, WI 53223 USA; shareware versions are available on many computer bulletin boards.) If I had archived the pollen files as T50RAW.ZIP, the file names and a lot of information about them, could be printed on the screen using the command PKZIP /v T50RAW because that ability is built into PKZIP. You can do the same thing with a self-extracting file if you specify the extension: PKZIP /v T50RAW.EXE. PKUNZIP.EXE does not recognize the /v switch, and the self-extracting zipped file is a subset of PKUNZIP. Many users believe the only thing they can do with a self-extracting file is to type its name and have the whole lot unzipped at once. There is another way, and I will use T50RAW.EXE as an example. You can find out the site file names by typing T50RAW /t which causes the program to test its files. A proprietary statement appears on the screen followed by a test of each compressed file by name--which gets an OK if it passes. You can redirect this list to a disk file by typing T50RAW /T > T50SITES.TXT and read it at your leisure. You can also extract one or a few files. The command T50RAW e*.raw p*.raw will extract all the files with extension "RAW" that begin with "E" and "P". This works with self-extracting files made with the full-strength [*p.16 / p.17*] registered version of ZIP2EXE, not the "junior" versions. Reference. Webb. R.S., Overpeck, J.T., Anderson, D.M., and Bauer, B.A. 1993. A description of the NOAA Paleoclimatology Program and World Data Center-A for paleoclimatology. INQUA - Commission for the Study of the Holocene, Working Group on Data-Handling Methods Newsletter 9:10-12. CANPLOT: A FORTRAN PROGRAM FOR PLOTTING POLLEN (AND OTHER) DATA. Ian Campbell Forestry Canada, 5320-122 St. Edmonton, Alberta, T6H 3S5 E-mail: icamp@nofc.forestry.ca CANPLOT is a program for creating pollen diagrams of either the sawtooth or the bargraph variety. In its current incarnation, it takes raw counts (the data format is Minnesota Compact) and control codes in as a single combined input file, and outputs PostScript code. PostScript is a page description language available on many laser printers and other output devices, or separately through a PostScript interpreter such as GoScript from LaserGo Inc. (9369 Carroll Park Drive, Suite A, San Diego, California 92121 USA) which I purchased recently at about CAN$130 or US$100. The advantages of PostScript output are discussed by Keith Bennett (1992). They are essentially that the same output file can be reproduced on any output device equipped for PostScript without modification, and it can be stored and transmitted as an ASCII file. It is also possible for a user to rapidly gain sufficient familiarity with the language to make minor editorial changes directly in the PostScript code, which I do quite routinely. For instance, Fig. 1 attached to this article was originally formatted for 8.5 X 11, and re- sized by Lou using the directions below: 1) determine the reduction ratio as you would for a reducing photocopier (i.e., desired dimension/original dimension). 2) at the beginning of the PostScript code, after the lines starting with % or /, find the first actual command - in CANPLOT files, the command INI on a line by itself. 3) after this command, insert the line: ".8 .8 scale" where .8 is the 80% reproduction ratio, and "scale" is written in lowercase only. 4) send the file to the printer. 5) if a distorted reduction is needed (i.e., different scaling in the X and Y dimensions), adjust the numbers accordingly - the first number scales the X-axis, and the second scales the Y-axis. ... Fig. 1. CANPLOT PostScript plot re-sized from file received by e-mail. In its present version, CANPLOT will allow a number of options in the diagram (see Campbell and McAndrews, 1992). Additional options not in the general release version (but may be made available by special arrangement, and at your own risk) include automatic plotting of temperature inferred from any one of a large set of published pollen-climate transfer functions for eastern North America. Other options available in the release version include user-selected vertical and horizontal scaling, zoning [*p.17 / p.18*] with various shades of grey, the automatic application of R-values and Y-intercepts to produce estimated biomass diagrams, user-selected exaggeration curve, vertical scaling by depth or by interpolated date, comments in the right or left margins which can be bound to specific levels, and stratigraphically constrained cluster analysis following Grimm (1987). Up to 250 levels and 250 taxa may occur in the same diagram. If the diagram is too long for a single page (both letter and legal size paper is allowed) then page breaks are inserted between curves, never in a curve, so that splicing the pages together is not a problem. The input data files can be created handily enough using any ASCII text editor, but a very user-unfriendly input module is available for those with a masochistic streak. This input module includes one handy feature, the automatic calculation of percentages of organic material, carbonates, and ash when using dry ashing for loss on ignition. The entire program for CANPLOT is published in the Computers & Geosciences paper, or can be had, together with the entry module and a small user's manual, from Ian Campbell, Forestry Canada, 5320 122 St., Edmonton, Alberta, T6H-3S5; icamp@nofc.forestry.ca; (403)-435-7300. References. Bennett, K. D. 1992. Use of PostScript to increase portability of pollen diagrams. INQUA - Commission for the Study of the Holocene, Working Group on Data-Handling Methods Newsletter 7: 6-7. Campbell, I. D. and McAndrews, J. H., 1992, CANPLOT: a FORTRAN-77 program for plotting stratigraphic data on a Postscript device. Computers & Geosciences 18: 309-335. Grimm, E. C. 1987, CONISS: a FORTRAN-77 program for stratigraphically constrained cluster analysis by the method of incremental sum of squares. Computers & Geosciences 13: 13-35. A SHORT INTRODUCTION TO PostScript CODE Ian Campbell Forestry Canada, 5320-122 St. Edmonton, Alberta, T6H 3S5 E-mail: icamp@nofc.forestry.ca PostScript is a page description language available on a wide variety hardware products, most notably laser printers and linotype machines (see Bennett 1992, and Campbell, this issue). Although PostScript is in ASCII format, it can be quite inscrutable and frustrating. This short article is not meant as a full-scale introduction to the PostScript language, but rather as an ice-breaker for those interested more in modifying PostScript graphics rather than writing their own PostScript applications. As such, it includes only a small subset of the PostScript language. A PostScript file normally starts with a %! which is usually followed on the same line by some information about the driver which created the file. The % symbol is used at the beginning of any line to start a non-executed comment. %% is used to indicate a comment of particular importance, but this is conven- tion rather than syntax. In the first few lines, a number of file-specific commands may be defined. PostScript allows the user to define new commands anywhere in the file, but they can be used only after they are defined, so definitions usually come first. Definitions can be made in two ways: /RM /rmoveto load def defines RM to mean "move the cursor X units to the right and Y units up from its current location, wherever that may be"; the command would be used by preceding the RM with the numbers of X and Y units desired, as in: 3 6 RM Longer definitions are contained in parentheses: /RS {currentpoint gsave translate rotate} def This loads the commands "currentpoint gsave translate rotate" in a particular stack, and defines RS as being equivalent to that set of commands. The new command would be used as in: [*p.18 / p.19*] 45 RS which would have the effect of rotating (rotate) the universe 45 degrees using the current cursor location (currentpoint ... translate) as the axis of rotation; first, however, the current orientation of the universe and all other parameters of the graphics universe are saved for later restoration (gsave). By now you have the idea that commands, whether user-defined or original PostScript, are preceded by the necessary parameters - as the command "def" is preceded by the definition, and cursor movement commands are preceded by the coordinates required. The most commonly used original PostScript graphics commands are: gsave : saves the graphic state for later restoration. This includes things like orientation of the universe and scaling. grestore : restores the saved graphics state. x setlinewidth : sets the width of a line to x units. x setgray : sets the color to a device-defaulted halftone of x% white (0 is black, 1 is white). newpath : starts a new virtual line for later manipulation. currentpoint : returns the X,Y coordinates of the current cursor location. These values can be used to feed into another command. x y moveto : moves the cursor to location x,y dx dy rmoveto : moves the cursor dx units to the right and dy units up from the current location. x y lineto : moves the cursor to x,y and draws a virtual line on the way. dx dy rlineto : moves the cursor dx units to the right and dy units up from the current location, and draws a virtual line on the way. closepath : closes a polygon by drawing a virtual line from the cursor to the start of the virtual path (usually the location where the newpath was started). clip : uses the current path as the clipping region - i.e., the actual printing window. The clipping region is defaulted at the start to infinity, but is in practice limited by the output device. Most laser printers will not print closer than about 4 mm to the edge of the page. The clippath can be restored by preceding the clippath command with a gsave command; the grestore command will restore the clipping path in effect at the time of the gsave. fill : fills a region defined by a virtual line with the current shade of gray. stroke : traces the virtual line with the current shade of gray. The virtual line is not to be confused with the path; the path may include segments (created with moveto or rmoveto) which are not part of the virtual line (created with lineto or rlineto), so that the virtual line which is to be stroked may include disjunct segments. x y scale : re-sizes the X and Y axis units by multiplication by x and y respectively (eg, ".8 .5 scale" will reduce the vertical axis to 50% of original, and the horizontal axis to 80% of original). The units are initially 1/72 inch. x y translate : moves the origin to location x,y. The origin is defaulted to the bottom left corner of an upright page. x rotate : rotates the universe x degrees counter-clockwise. Other commonly used commands are: showpage : instructs the printer to print the page which has just been described. Any further commands will refer to the next page. tttttt findfont : loads the font ttttttt into memory. Most PostScript devices have several fonts built into them which can be accessed by this command; the most commonly used are Times-Roman, Helvetica, Times-Italic, Helvetica-Oblique, TimesBold, Helvetica-Bold, Courier, Courier-Oblique, and Courier-Bold. tttttt setfont : makes tttttt the font the one to use. x scalefont : scales the font so that a capital A is x units high. [*p.19 / p.20*] (string) show : prints the string in brackets in the current font, starting at the current cursor location, and extending along the current orientation of the X-axis. A short PostScript program you can try, using only those commands discussed above, is this: %! Handwritten PostScript file. /INI {.24 .24 scale 1200 1500 translate 1 setlinewidth 0 0 moveto} def /BOX {600 0 rlineto 0 600 rlineto -600 0 rlineto 0 -600 rlineto} def /TWIST {9 rotate 0 0 moveto} def /DRAW {BOX gsave fill grestore 0 setgray stroke TWIST} def /SG /setgray load def /SF /Times-Roman findfont def /H {SF 60 scalefont setfont} def %% page 1 of 1 INI 0 SG DRAW .1 SG DRAW .2 SG DRAW .3 SG DRAW .4 SG DRAW .5 SG DRAW .6 SG DRAW .7 SG DRAW .8 SG DRAW .9 SG DRAW 1 SG DRAW -99 rotate -400 300 moveto .5 setgray H (PostScript!) show showpage There are also math operators, boolean operators, and logical operators like "if", but that is going too far for this ice-breaker. The main thing to remember is that most devices, when faced with a PostScript error, will simply spin their electrons for a while then give up. They may not give you any indication of an actual error in the PostScript code, so make modifications carefully and try to test them with a software PostScript interpreter before trying to print the file. If you are interested in learning more about PostScript, the best references are (and they really go together): PostScript Language Reference Manual, and PostScript Language Tutorial and Cookbook, both from Addison-Wesley. They are well-written, well-organized reference volumes, and are available at a reasonable price (I paid $56.50 Canadian for both). Reference. Bennett, K. D. 1992. Use of PostScript to increase portability of pollen diagrams. INQUA - Commission for the Study of the Holocene, Working Group on Data-Handling Methods Newsletter 7: 6-7. MVSP 2.1 Warren L. Kovach Kovach Computing Services 85 Nant-y-Felin Pentraeth, Anglesey LL75 8UY Wales, UK E-mail: warrenk@cix.compulink.co.uk I have recently released a new version of my program MVSP - A MultiVariate Statistical Package for the IBM-PC and compatibles. For those not familiar with it, MVSP performs a variety of ordination and cluster analyses. It provides an inexpensive yet easy means of analysing your data in fields ranging from ecology and geology to sociology and market research. MVSP is in use at hundreds of sites in over 45 countries. The results of analyses using MVSP have been published in numerous journals, including Science, Nature, Ecology, Journal of Petroleum Geology, and Journal of Biogeography. MVSP version 2.1 has several improvements over version 2.0: 1) Detrended correspondence analysis 2) Built-in dendrogram production - drawn automatically in either graphic or text mode 3) Improved printer support - can now output text and/or graphics on Epson dot-matrix, IBM Proprinter, HP Laserjet & Deskjet, HP Paintjet, and postscript laser printers, HP plotters, and .PCX bitmap files. 4) Data import and export - transfer data between MVSP and the following file formats: Lotus 1-2-3 and Symphony, Cornell Ecology Programs. 5) Protected mode version - directly access up to 16 mb of RAM memory, not just 640 k. Eliminates swapping to disk and dramatically speeds up analyses of large data sets. Only for computers with 80286 or higher microprocessors. 6) Save ordination scores to Lotus files - for further analysis or graphing. 7) Execute DOS commands - from the MVSP menu 8) Several improvements in the user interface & bug fixes MVSP costs 65 UK pounds or US$100. A limited educational version (limited to 100 x 100 matrices) is [*p.20 / p.21*] also available at attractive site license prices (ranging from 5-10 UK pounds per copy; US$8-16). Contact me for more details. Tilia AND Tiliaùgraph 2.0 A major upgrade to Tilia and Tiliaùgraph will be available in a few months. All present owners of versions 1.x will be notified when the upgrades are available. In addition to a variety of new features, these versions are designed to work with the North American and European Pollen Databases. Some of the new features are: * 8-characater VarCodes rather than 2-character * 60-character VarNames * Multiple age and depth variables permitted * Null data values (Permits plotting data with different depths on the same diagram.) * Large spreadsheets can be stored in extended memory * Forms (Allows storing of data such as site details, radiocarbon dates, lithologic descriptions, etc. A set of standard forms are provided, or you can design your own.) * More input/output options * More complete user manual * Lithology symbols * Pollen counting program PCOUNT is included * Graphical display of age/depth graphs 8 Default style for pollen diagram can be set by user A NEW VERSION OF PCSLOT Malcolm Clark Department of Mathematics Monash University Clayton, Victoria Australia 3168 E-mail: rmc@monu1.cc.monash.edu.au I wish to announce that a new version of PCSLOT (Clark, 1992, 1993), my PC-based program for sequence-slotting based on the CPL criterion, is now available. Version 1.5 has two important new features. Firstly, users may specify the Chord Distance (useful for pollen sequences) as the distance measure to be used. Secondly, users with access to a PostScript laser printer can produce a high-resolution half-tone "picture" of the H-matrix, as a summary of the slotting process. This option is particularly useful in cases where both sequences are long (e.g. more than 70 levels), because it ensures that the H-matrix is printed on a single page. ... Fig. 1. H-matrix via PostScript and e-mail. Fig. 1 shows the picture produced by PCSLOT Version 1.5 of the H-matrix in Clark (1993, fig. 2, p. 8). The "river" of asterisks defining the alternative optimal slottings now becomes a black band running more-or-less diagonally across the picture. Elsewhere, higher H-values show up darker than lower H-values. For example, compare the bottom-left and bottom-right corners of the two diagrams. The dark region in the bottom-left corner corresponds to the block of high H-values in the bottom-left corner, indicated by R's in the previous diagram. Conversely, the light area on the bottom right matches the region of A's (low H-values). Version 1.5 is available in the file PCSLOT15.EXE, obtainable by anonymous ftp on the Internet (ftp geology.wisc.edu or ftp 144.92.137.14 from the /pub/inqua directory. This self-extracting file contains help files, test data and all necessary .EXE files. [*p.21 / p.22*] Reference. Clark, R. M. 1992. Sequence comparisons and sequence-slotting. INQUA - Commission for the Study of the Holocene, Working Group on Data-Handling Methods Newsletter 8: 3-6. Clark, R. M. 1993. Assessment of sequence-slotting. INQUA - Commission for the Study of the Holocene, Working Group on Data-Handling Methods Newsletter 9: 5-10. A POLLEN DATABASE BULLETIN BOARD? Konrad Gajewski Department of Geography University of Ottawa 165 Waller Street Ottawa, Ontario Canada K1N 6N5 E-mail: gajewski@acadvm1.uottawa.ca Now that Tom Webb's pollen database is publicly available, we can all pack away our microscopes and spend the rest of our time at our terminals. However, those who have used previous versions of this database know of the large number of questions that arise when you try to use it. Figuring out how to get the data we need from the Illinois group's PARADOX files and dealing with the numerous problems and errors that are surely in the database could take up several years of our lives. Similarly, users of noncommercial software like TILIA can be frequently frustrated when it does not do what (we think) it should, and Eric or John are out of town or not answering their electronic mail. One way to exchange information quickly is a "pollen data-handling" bulletin board on the Internet. In such a bulletin board, we could contact each other to get answers to problems with the various software that we all use. It is a quick way to get details of data in the file, questions to the person who did the study about possible errors in the file, questions about supplemental information that might be missing, etc. It could be a place for more general information exchange - including new laboratory techniques, coring techniques, new papers we have published. In short, the kind of information included in this newsletter, but that could arrive in your office every day. When Lou and I talked about this last spring, I contacted Henri Grissamo-Meyer, who is the person running a tree-ring bulletin board on a day-to-day basis. He sent me a long description of how a bulletin board operates, and seems quite willing to help in the establishment of a palyno board. The "listserver" software to do this is widely available, and is supposedly documented at every University. The problem is that I do not have the time to manage it, and I suspect that many others do not either. It does take an hour or two each week. Is anyone interested in doing this? If so, I can send along Henri's message, or pass on his coordinates. He has suggested that it is a nice job for a graduate student, as it is a way quickly to know everyone in the business. Presumably, it should be run from a larger permanent lab or research institute to ensure some long-term continuity. Anyone interested? POLPAL-L, A LIST-SERVER FOR PALYNOLOGISTS Louis J. Maher David Green recently called my attention to the list server POLPAL-L dealing with palynological topics that Peter Kevan set up at the University of Guelph. This is perhaps not the same as the bulletin board idea that Konrad Gajewski mentions above, but it has some of the same purposes. Once you join the list, any e-mail you send to polpal-l@vm.uoguelph.ca is relayed on to everyone on the list. You can get OFF the list at any time by following the instructions you receive on joining. From the Internet, you can join POLPAL-L by sending to listserv@vm.uoguelph.ca an e-mail message which consists entirely of the single phrase: SUBSCRIBE POLPAL-L Your Name Your Name should be only your given name(s); the listserver automatically extracts your e-mail address from your message header. You will soon get back a message somewhat like the much-shortened version that follows: [*p.22 / p.23*] Dear networker, Your subscription to list POLPAL-L (POLPAL-L Discussion List) has been accepted. You may leave the list at any time by sending a "SIGNOFF POLPAL-L" command to listserv@vm.uoguelph.ca. Please note that this command must NOT be sent to the LIST address (POLPAL-L@vm.uoguelph.ca) but rather to the LISTSERV address (listserv@vm.uoguelph.ca) ..... More information on LISTSERV commands can be found in the LISTSERV reference card, which you can retrieve by sending an "INFO REFCARD" command to listserv@vm.uoguelph.ca. Virtually, The LISTSERV management If you are on Bitnet, you should substitute the following addresses for those I used above: polpal-l@_ uoguelph.bitnet and listserv@uoguelph.bitnet. One can get a list of the e-mail addresses of members who have not opted to conceal themselves. I noted several names who get the Data-Handling Newsletter. Lately most of the communications seem to be dealing with "eddy viscosity" and "eddy diffusivity." A couple of members tried to quit, but sent the request to POLPAL-L@uv.uoguelph.ca instead of listserv@um.uoguelph.ca, so evidence of their cowardice went out to everyone! POLPAL-L looks like a very good idea, and I have been going to ask about Macintosh programs that will plot Quaternary-style pollen diagrams (a couple are in the works) and what to do about my degrading reference collection stored in silicone. HYPERMEDIA AND PALAEOENVIRONMENTAL RESEARCH David G. Green Bioinformatics Facility Research School of Biological Sciences Australian National University GPO Box 4 Canberra 2601 AUSTRALIA E-mail: david.green@anu.edu.au The growth of communication via international information networks is revolutionizing the way in which research is done (Green, 1993). The impact of the Gopher protocol over the last eighteen months has been phenomenal. However, Gopher menus are far from the end of the story. We are now seeing the beginning of a proliferation of hypermedia informa- tion services via the World Wide Web. Taken together these new protocols offer completely new ways of compiling, analyzing and disseminating scientific information. Here I briefly describe this coming hypermedia revolution and outline some of the implications for the ways in which we handle data. ... Fig. 1. Hypermedia information. Here Xmosaic is used to access documents (including in-lined images) from ANU's World Wide Web pollen interface. Highlighted terms are hypertext links to other documents or images. Hypermedia information. The term "hypermedia" refers to the combination of hypertext and multimedia. A hypertext document is one that includes links to other documents. To a user hypertext links appear as highlighted numbers, terms or icons (pictures that indicate links). Selecting an icon, or highlighted text, causes the system to retrieve the indicated file, from wherever in the world it is stored. "Multimedia" refers to information that combines a variety of forms, including text, graphics, sound, and even animation. The World Wide Web (WWW), developed at CERN (Switzerland), uses the HyperText Transfer Protocol (HTTP) to distribute hypermedia information across [*p.23 / p.24*] Internet (Krol, 1992). Like Gopher, WWW operates via client-server transactions. That is, the user's client software transmits a request to a server, which then transmits back the requested information. Rather than menus (which Gopher uses), WWW deals with hypertext documents, which contain embedded links. WWW is potentially far richer and more flexible than Gopher. For example, WWW can combine help files, images and menus all in a single document. World Wide Web has not yet achieved the popularity of Gopher because adequate client software has not been available. However this has now changed. Early in 1993 the National Centre for Supercomputing Applications (NCSA) released a new WWW browser (Xmosaic) for workstations (Fig. 1). Xmosaic not only provides an easy-to-use interface for WWW, but also for every other major network protocol (including FTP, Mail, WAIS and Gopher). Versions of Mosaic for PCs (DOS and Mac) are due for release late in 1993. A beta version of Cello, a WWW client for MS Windows is already available along with test version of clients for Macintosh and several other platforms (Table 1). Several sites provide text-based access to the Web, via Telnet or Gopher (Table 2), for users who do not themselves have the necessary client software. An important feature of Xmosaic is that it can also be used locally, without recourse either to Internet or a server, as a user front-end for browsing hypermedia information (Fig. 2). It thus provides a completely seamless transition between a user's own, local information and network information. Xmosaic also provides hooks for linking it to other software. It could, for instance, be used as a browser for on-line help or viewing data. Also, unlike other protocols, HTTP can provide formatting information via the HyperText Markup Language (HTML) (Fig. 2). Mosaic, and other browsers now appearing, thus turn WWW documents, into electronic books. Taking advantage of the new technology, electronic journals, based on Internet, are now appearing with increasing frequency. Hypertext links in HTML take the form of a "Uniform Resource Locator" (URL), which specifies not only WHERE to find a document, but also HOW. The example (Fig. 1) shows how hypertext links appear to a user (the underlined text) and how they are specified via URLs (Fig. 2). Hypermedia services. At present the number of World Wide Web servers is small. Many servers are Table 1. Sources of client software --------------------------------------------- Name Platform Site/Path --------------------------------------------- World Wide Web clients ---------------------- Various Many! info.cern.ch /pub/www MacWWW Mac info.cern.ch /pub/www/mac cello DOS Windows fatty.law.cornell.edu /pub/LII/Cello xmosaic X11 Work- ftp.ncsa.uiuc.edu stations /Mosaic Gopher ------ xgopher X11 Work- boombox.micro.umn.edu stations /pub/gopher turbogopher Mac wingopher DOS Windows ...etc Many others! Table 2. Telnet access to Gopher and World Wide Web (1) ------------------------------------------ Gopher ------ telnet info.anu.edu.au (Australia) telnet gopher.th-darmstadt.de (Germany) telnet gopher.uiuc.edu (USA) Login name: gopher or info Password: not needed World Wide Web (2) -------------- telnet info.cern.ch (SWISS) telnet FATTY.LAW.CORNELL.EDU (USA) telnet vms.huji.ac.il (ISRAEL) Login name: www Password: not needed --------------------------------------------- Notes: (1) These services are provided for people who do not themselves have client software for Gopher or World Wide Web. They are text only. For a more comprehensive list of sites and services see ftp: life.anu.edu.au/bioinformation/faq/_ internet_services (2) These services indicate hypertext links via numbers inserted in the text, e.g. (21). Typing the relevant number at the prompt selects the document concerned. little more than gateways to a site's FTP or Gopher servers. At the Australian National University the on-line hypermedia information includes such diverse material as an illustrated history of art, illustrated tutorials on fractals and demonstrations of scientific visualization. The Bioinformatics Facility provides hypertext "home pages" for many of its services, [*p.24 / p.25*] including biodiversity, biomathematics, global monitoring, landscape ecology, molecular biology and paleoenvironmental information. Markup and data handling. As mentioned above, WWW hypertext documents are marked up using HTML. HTML is an application of the Standard Generalized Markup Language (SGML), which is emerging as a leading international standard for preparing both hypertext and multimedia material for publication (Fig. 2). The reasons for the growing popularity of markup, rather than the formatting provided by common word-processors, is its flexibility. The same tags can be interpreted in different ways for different uses of the same material. The interpretation of tags is defined by an SGML Data Type Definition (DTD). This file, the equivalent of a word-processor style sheet, defines how software interprets and uses tags when processing a document. For instance an SGML-compatible word-processor might interpret the tag as a heading to be formatted accordingly, whereas an SGML-compatible database might interpret it as a field in a data record. The growing importance of markup has important implications for anyone preparing or handling data, and for those of us who are writing software. Some standard formats supported by Xmosaic are listed in Table 3. Formats. In designing data formats, we should use embedded tags to indicate records and fields in datasets. We should avoid idiosyncratic formats, especially those that require data to be laid out in a [*p.25 / p.26*] particular pattern. The use of tagged field formats has been standard practice on Internet for years (Green, 1993). More and more freeware programs available on Internet assume that data uses markup or tagged fields. There are many other advantages too. For instance, the order of data fields within a record is unimportant and default values can simply be omitted. Conversion of formats and database indexing (e.g. using waisindex) is greatly simplified. (1) A typical idiosyncratic format Curry Pond 16 40 0 10 20 30 40 (2) Simple tagged field format SITE Curry Pond SAMPLES 40 TAXA 16 DEPTHS 0 10 20 30 40 (3) Markup of text with embedded data The pollen record for Curry Pond in Nova Scotia includes counts for 16 taxa in 40 samples. Sediment samples were extracted at the following depths: 0 10 20 30 40 . Figure 3. A data fragment formatted in three ways. Markup means that data can even be embedded in documents and processing instructions can be included within datasets. In effect any document is treated as a program, with commands and data mixed together. Formatting the same fragment of data in three different ways (Fig. 3) illustrates the differences between the above formats. Distributed datasets. The use of URLs makes possible exciting new way of organizing data. For example, it is now possible, using URLs, to draw together, in a single document, links to all the datasets, documents and software relevant to a particular problem (cf. Fig. 2). For instance, an electronic paper or preprint, published on the Web, can include references to all the tools and information used in the study. Moreover unlike references in a paper journal, these references would allow the reader to retrieve the relevant data or software automatically. Also reviews of research on particular topics could include live links to all the datasets or documents involved, thus creating an instant database. Software. For those of us who write software, there are several advantages in incorporating tags and markup into our programs. On the one hand users are more likely to use a program if they do not have to recast their data in a special format in order to use it. Also providing programs that use standard network formats (Green, in press) encourages users to contribute their data to network databases and makes it possible for users to pool and use data, software and other resources from a variety of sources. Table 3. Some of the file formats used in World Wide Web --------------------------------------------------------- Text .txt ASCII (any unrecognized extension) Formatted text * .html HyperText Markup Language .dvi Data Visual Interchange .ps Postscript Images .hdf Hierarchical Data Format (NCSA) .jpg Joint Photographic Experts Group (JPEG) .gif Graphics Interchange Format (CompuServe Inc.) .rast Sun Raster Files .ras Sun Microsystems, Inc. .sgi Silicon Graphics Image Files .rgb Red/green/blue format .tiff Tagged Image File Format Files (Microsoft & Aldus) Audio .aiff Audio interchange format (silicon Graphics) .au SUN/NeXT sound format Animation .mpeg Movie format, Photographic Experts Group (MPEG) .movie Silicon GI audio/video movie --------------------------------------------------------- Note: * TeX and LaTeX files must be converted to DVI Conclusion. At the time of writing hypermedia information still suffers from several handicaps. The first, as noted above, is a lack of client software for PC's (DOS and Mac). The second problem is a lack of widely available WYSIWYG editors for SGML/HTML. So far I have done all markup either by hand, or else by writing scripts to perform specific conversions (e.g. Ventura tags to HTML tags). However this situation is changing rapidly. Recent versions of several popular word-processors include SGML macros and several conversion utilities are available. The final problem is the limitation of the HTML standard: it does not handle complex equations for instance. At present these must be cast in TeX/LaTeX separately and a processed for viewing by xdvi or a postscript viewer. [*p.26 / p.27*] References. Green, D.G. (1993). Databasing the world. INQUA - Commission for the Study of the Holocene, Working Group on Data-Handling Methods Newsletter 9:12-17. Green, D.G. (in press). Databasing diversity - a distributed, public-domain approach. Taxon, in press. Krol, E. (1992). The Whole Internet. O'Reilly and Associates. THE INQUA FILE BOUTIQUE Louis J. Maher The following files are currently in sub-directory /pub/inqua, and those on the Internet can pick them up free by anonymous ftp [ ftp geology.wisc.edu or ftp 144.92.137.14 ]. They are described in the ASCII file readme.txt. Name Bytes Format bot1intr.txt 13325 ASCII botbib1.txt 121456 ASCII botbib2.txt 55808 ASCII bta.exe 30630 Binary coresegz.exe 22713 Binary mvsp21.exe 310081 Binary newltr-1.txt to newltr10.txt ASCII palyplt.exe 148042 Binary pcslot15.exe 147412 Binary pol-ega.exe 682992 Binary pol-herc.exe 641528 Binary pol-vga.exe 693618 Binary pol-vgam.exe 649724 Binary polcntpk.exe 103426 Binary polfil15.exe 77393 Binary polish.exe 206061 Binary psimpolz.exe 117173 Binary rarefact.exe 125575 Binary readme.txt ASCII sltdepeg.exe 92702 Binary sltdepvg.exe 93575 Binary spoon-ez.exe 21920 Binary t50raw.exe 644039 Binary tranzip.exe 253476 Binary If you wish to add a useful file to the INQUA boutique, please put it in the subdirectory /pub/incoming, e-mail me saying what you have done, and tell me what description you want me to add to the boutique's readme.txt file. DATA-HANDLING WHILE SAMPLING CORES Louis J. Maher Sediment samples for palynology can be taken from cores in a tiny fraction of the time it takes to process and count the pollen. Because of that fact, about a dozen years ago I decided it would not prolong things too much if I made some extra measurements that would allow me to express the microfossil abundance either by volume or wet weight or dry weight, and at the same time record the water content of the sediment. If the samples were spiked with exotic pollen marker grains, pollen abundance could be expressed in units of grains/cm3, grains/g[wet] or grains/g[dry] as well as the traditional pollen percentage. And if the age of the sediment were known, one could also express the abundance in grains/cm2/yr (Maher, 1981). I worked out a procedure for packing sediment into a calibrated stainless- steel measuring spoon, leveling the sediment with a knife edge, weighing it, and splitting the sample into two portions--one of which was wrapped in aluminum foil and dried in an oven; the other was processed for its pollen. All the sediment quantities were based upon weighing the empty spoon and foil, and accounting for the volume of the pollen and oven sub-samples by their weight fraction of the total wet weight of the spoon's sediment. (This procedure works best when the sediment is fine-grained and water-saturated.) I shudder when recalling the effort it took to take 134 samples from the core segments. Each sample required six steps of taring and weighing which were done with an older model torsion balance that was difficult to use. If run by an experienced and stubborn individual, it was capable of an accuracy of + 0.002 g. As the core was being sampled, each reading was double-checked and recorded on a paper tablet. A blank was left for the oven-dried portion which was weighed the following day. The raw data were then manually entered into a programmable calculator for further processing. I had a lot of data, but it was not handled very efficiently. When I recently had occasion to take similar samples from a new series of cores, I decided it was time to purchase an electronic balance with a large digital display. In studying catalogs, I found that many of these instruments come with bi-directional RS-232-C [*p.27 / p.28*] interfaces--meaning they can be controlled from a computer. I decided a good choice for me was a Denver Instrument XE Series Model 100A sold by Fisher Scientific; it offered precision (+/- 0.0002 g) and simple operation at a very competitive price. Even though it had a large digital display, it seemed like a good idea to write a computer program so that the data could be acquired without ever having to copy any numbers by hand. The Model 100A's RS-232 port is like a socket on a US telephone which has four leads. Our electronics staff made a jack and cable to connect to the COM1 port on my IBM-compatible laptop computer. SPOON-EZ.BAS is a QuickBasic program. It establishes a communication link with the balance at 2400 baud, one of the standard BASIC rates that also works with the Model 100A. The program functions well with the QBasic interpreter that comes with DOS 5.0. I suspect that the inner workings of many of today's electronic balances utilize very similar computer chips and that their commands are much alike. When the communication link is established between computer and balance, the usual "hand-shake" does not take place. The statement: OPEN "COM1:2400, N, 8, 2, RS, DS0" FOR RANDOM AS #7 LEN = 256 uses the RS and DS0 switches to tell the computer it should not "time out" when it receives no acknowlegement that the connection was opened. Let me explain the program by showing some sample data, and then I will explain how you can get a free copy and adapt it to your own needs and equipment. It is assumed that the core will be unwrapped and ready for taking samples-- say from top to bottom--and that a sample's depth will serve as its label. You have a calibrated spoon and a supply of aluminum foil squares about 5 cm on a side that you have cut from a roll of foil. The foil will hold the sub- sample to be dried in the oven; its sample number can be engraved on a corner with a ballpoint pen. The spoon will be cleaned and dried after each use. I give it a moment in a ultrasonic bath with a little detergent, and then dip it in a small beaker of alcohol. When the spoon is then touched to a flame, the alcohol burns off quickly, leaving the spoon clean and dry for the next sample. Generally the core sampling will take some time and may extend to a second day. One should therefore be able to stop and save data to a file, which can be recovered and continued at a later time. The initial weight file (with extension *.IWT for Initial WeighT) will have seven categories (1. Serial Number, 2. Empty Spoon Weight, 3. Spoon + Sediment Weight, 4. Empty Foil Weight, 5. Foil w/Sediment for Oven Weight, 6. Spoon w/Pollen Sample Weight, and 7. Foil w/Sed FROM Oven). The seventh category will be unknown (stored as a zero) until the oven-dried sample is weighed at a later session. When the program is run, it asks questions ("Does a *.IWT file exist for this core?", "Do you wish to do another sample? (Y/N)", etc.), and issues instructions ("Touch the Space Key to Tare the Empty Balance", "Touch the Space Key to weigh the clean empty spoon", "Touch the Space Key to weigh the filled spoon with its leveled sediment", etc.). Each time a weight is needed, the program actually instructs the balance to send 10 separate measurements, and the mean value is the one recorded. The balance is designed to provide a reading only when it meets its own standard for stability. The computer beeps whenever it is ready for a new measurement. When the oven-dried samples are ready, one should be able to load the stored *.IWT file to complete the last column. The program will call for the aluminum-wrapped samples by sample number, and the dried sample's weight will also be the mean of 10 readings. This part of the process does not take long so there is no provision for saving the file in mid task. When all the samples in the file are complete, a Final WeighT file (*.FWT) is saved; it will contain all the information needed to characterize the pollen samples' water contents, weights, and volumes. Spoon-EZ.BAS by L. J. Maher, Version 1.8 PROGRAM TO AUTOMATE WEIGHING SPOON SAMPLES FOR POLLEN CONCENTRATION ---------------------------------------------- 1. RECORD RAW WEIGHTS OF SAMPLES FROM CORE TO AN INITIAL DISK FILE [You can save the Initial WeighT (*.IWT) file, stop, and continue it later.] 2. WEIGH DRIED SAMPLES FROM OVEN TO FINISH INITIAL DISK FILE [Weigh ALL the Dried Samples in the (*.IWT) File in ONE operation. The results will be saved as a Final WeighT (*.FWT) File] (Steps 1 and 2 MUST BE COMPLETED BEFORE steps 3 and 4.) 3. PRODUCE COMPREHENSIVE DISK FILE OF POLLEN SAMPLE VOLUMES AND WEIGHTS 4. PRODUCE MINI DISK FILE OF JUST SAMPLE VOLUMES AND WEIGHTS E. End Program Press 1,2,3,4 or E to continue _ Fig. 1. Main menu of SPOON-EZ.BAS [*p.28 / p.29*] Fig. 1 shows the initial screen menu. Pressing "1" allows you to make an Initial Weight file. Pressing "2" allows you to finish processing the *IWT file, converting it into the Final Weight file. Obviously, a *.FWT file must be available before going to steps "3" or "4". Whenever you press "1" or "2" to begin either sequence of measurements, the program asks you to press a key to tare the empty balance pan. You are then instructed to put a 50 g scale weight on the pan and press a key so that the balance will calibrate itself. It is probably good practice to weigh the same 50 g object again at the end of a long session, but the program does not require it. I thought it would help to include some simple test files here to show the results. For the test, I have used my actual measuring spoon and five separate pieces of aluminum foil. The stainless spoon was sold as a 1/4 teaspoon measure. A dictionary conversion chart suggests that a level 1/4 teaspoon should equal 1.232 ml. I calibrated my spoon by weighing 100 leveled spoons of modeling clay of known specific gravity. The weight variation among the samples was equivalent to a mean spoon volume of 1.169 ml with a standard deviation of 0.014 ml. The calibrated mean volume of my spoon is therefore about 5 per cent less than the value stamped on its surface. In a normal run, no effort is made to keep the size of the oven sub-samples and the pollen sub-samples exactly constant; it is only important to know what they weighed. In this example, however, I will substitute gram balance weights for the sediment so that we can predict what the answers will be and then check them against the calculated results. I used a 3-gram and a 2-gram weight--5 g in all--to represent the weight of a level spoonful of sediment. The 3 g weight would be the pollen sample; the 2 g weight represented the por- tion of wet sediment that was going to be sent to the oven. The oven sample was placed on its foil and weighed. The pollen sample represents 3/5 of the mass (and therefore the volume) of the spoon which suggests--if the 5 g had actually been sediment that filled the spoon--that the pollen sample's volume will be 0.701 ml with s.d. 0.008 ml. The five test samples were given arbitrary labels. They were saved as INQUATST.IWT, which has the structure of a "Wisconsin RAW" file. It is duplicated below: Test of Known Weights 7 5 10 5.06846 10.068 .149 2.1491 8.0682 0 20 5.06849 10.06837 .09799 2.09827 8.06839 0 30 5.0678 10.06797 .1028 2.10316 8.068151 0 40 5.068379 10.06771 7.895999E-02 2.07877 8.068399 0 50 5.068389 10.0681 .15851 2.15823 8.06797 0 Serial Nbr Empty Spoon Wt Spoon + Sediment Wt Empty Foil Wt Foil w/Sed for Oven Wt Spoon w/Pollen Sample Wt Foil w/Sed FROM Oven This file made with Spoon-EZ.BAS, Version 1.8 on 06-28-1993 at 14:26:29. The first line is a title, the second line is the number of categories (7), and the third line is the number of samples (here 5). Data for the five samples then occur on succeeding lines, each with seven categories. The category names are added after the data, and a last line tells when the file was made. You will note that the last category--Foil w/Sed FROM Oven--is always zero in the *.IWT file. No effort is made to format the weight data; they are simply the mean value of the 10 readings. Generally, the only time a *.IWT file is ever looked at is when catastrophe strikes during a session-- such as dropping a sample on the floor. In that case, one can simply finish the sample's weighing sequence by recording the weight of an empty balance. When the program asks for the next sample number, enter the same number and re-sample the core at the same level. After the file has been saved, read it with as ASCII text editor, delete the bad line, and reduce the recorded number of samples (3d line) by one. Item 2 from the SPOON-EZ menu was then selected and INQUATST.IWT was loaded from disk. In this trial the oven-dry sample used the same foil as the original "wet" sample, but with a 1 g weight substituted. This simulates a sediment that has lost half its mass on drying; its (dry weight/wet weight) ratio is thus 0.5. When the five oven-dry samples were completed, the final weight file was saved as INQUATST.FWT which is reproduced below: Test of Known Weights 7 5 10 5.06846 10.068 .149 2.1491 8.0682 1.149 20 5.06849 10.06837 .09799 2.09827 8.06839 1.09854 30 5.0678 10.06797 .1028 2.10316 8.068151 1.10297 40 5.068379 10.06771 .0789 2.07877 8.068399 1.07881 50 5.068389 10.0681 .15851 2.15823 8.06797 1.15909 Serial Nbr Empty Spoon Wt Spoon + Sediment Wt Empty Foil Wt Foil w/Sed for Oven Wt Spoon w/Pollen Sample Wt Foil w/Sed FROM Oven This file made with Spoon-EZ.BAS, Version 1.8 on 06-28-1993 at 14:31:42. The Final Weight (*.FWT) file provides all the information needed to characterize the volume and [*p.29 / p.30*] weight of the pollen samples as well as their water content. Once you have made a *.FWT file, you can delete the Initial Weight (*.IWT) file. Now if you choose item 3 on the SPOON-EZ menu, you can produce a Volume-WeighT (*.VWT) file which displays all the data you would need to document your work. Part of INQUATST.VWT--showing sample 10 only to conserve space--is reproduced below: Test of Known Weights Number of samples = 5 This file made with Spoon-EZ.BAS, Version 1.8 on 06-28-1993 at 14:51:21. All weights based on mean of 10 balance readings. **SAMPLE NUMBER 10 Spoon Vol Mean = 1.169 ml S.D.= 0.014 ml Spoon Weight = 5.06846 g Spoon Plus Sediment Weight = 10.06800 g Empty Foil Weight = 0.14900 g Foil Plus Sediment For Oven = 2.14910 g Foil + Sediment FROM OVEN = 1.14900 g Total WET Sed Wt in Spoon = 4.99954 g Wt of Wet Sediment For Oven = 2.00010 g Wt of Dry Sediment FROM Oven = 1.00000 g Ratio of Sediment's Dry Wt to Wet Wt = 0.49997 Ratio Pollen Sed Wet Wt/Total Sed Wet Wt = 0.60000 Pollen Sample WET Wt = 2.9997 g Pollen Sample DRY Wt = 1.4998 g POLLEN SAMPLE VOLUME = 0.7014 ml S.D.= 0.008 ml **SAMPLE NUMBER 10 ... It is good practice to load an *.VWT file into your word-processor and run off a paper copy for your files. Finally, if you choose menu item 4, you can produce a SUMmary (*.SUM) file. And if you never want to copy any of these numbers by hand, you can load a *.SUM file, read the first 8 lines as dummy variables, and have each sample's values available to be read by machine for further processing. INQUATST.SUM is reproduced below: Test of Known Weights Number of samples = 5 This file made with Spoon-EZ.BAS, Version 1.8 on 06-28-1993 at 14:51:33. All weights based on mean of 10 balance readings. Sample Volume StdDev. Wet Wt Dry Wt Dry Wt/Wet Wt Nbr ml ml g g Ratio 10.0 0.701 0.008 2.9997 1.4998 0.49997 20.0 0.701 0.008 2.9999 1.5006 0.50020 30.0 0.701 0.008 3.0004 1.5002 0.49999 40.0 0.701 0.008 3.0000 1.4999 0.49997 50.0 0.701 0.008 2.9996 1.5009 0.50036 By examining these test data, it can be seen that the expected pollen sample volumes and standard deviations were recovered. The pollen sample wet and dry weights and the (dry weight/wet weight) ratios are accurate to about +/- 1 mg. If this program would be useful to you, it is available by anonymous ftp in the subdirectory /pub/inqua at geology.wisc.edu. SPOON-EZ.EXE is a self- extracting file containing SPOON-EZ.BAS and a sample *.FWT file. Menu items 1 and 2 will not work unless your computer's COM1 port is connected to a compatible electronic balance; you can try menu items 3 and 4 reading the *.FWT file in your default directory. Near the beginning of the BASIC source code, you will find a number of variables defined that can be changed to meet your own needs. These include the mean and s.d. of your measuring spoon, the name and model of your balance, the number of readings to average, and the commands to cause the balance to tare, calibrate and to send its weight reading. You can even change the name of the program; if you do, remember to rename the *.BAS file to conform to your choice. Consult the instruction manual for your electronic balance to determine how to adjust its baud rate. Most have set-up programs that can be accessed with a terminal emulator like PROCOMM. Measuring mud is [almost] fun again. Let me know if you have questions or problems. Reference. Maher, Louis J., Jr. 1981. Statistics for microfossil concentration measurements employing samples spiked with marker grains: Review of Palaeobotany and Palynology, 32: 153-191. PROGRAM SOURCES Owen K. Davis Department of Geosciences University of Arizona Tucson, AZ 85721 E-mail: palynolo@ccit.arizona.edu. The following information is excerpted from a table in a chapter entitled "Data Organization and Computers," by J.K. Lentin, O.K. Davis, K. Piel, and T. Munsey, which will appear in PALYNOLOGY AND STRATIGRAPHY, edited by J. Jansonius and D.C. McGregor, to be published next year by AASP Press. The information is primarily from past issues of this (INQUA Working Group on Data-Handling Methods) [*p.30 / p.31 *] newsletter. The addresses and information for obtaining copies have been brought up to date to the best of my and Lou Maher's knowl- edge. Please let me know of any errors. Hopefully, having all the citations together in one place will be helpful. DIAGRAM-PLOTTING PROGRAMS CANPLOT þ A program for drafting pollen diagrams that produces different curve styles, shading, and dendrograms in batch mode for PostScript þ printers. Price $10.00, contact Ian D. Campbell, Forestry Canada Northwest Region, 5320-120 ST, Edmonton, Alberta, Canada T6H 5A1, icampbell@nofc.forestry.ca. PALYPLOT þ A program for drafting pollen diagrams on IBM-compatible computers. Written by Craig A. Chumbley, it produces different curve styles, and shading on various plotters and printers by writing a file read by Generic Cadd Level 3, Generic Cadd v. 5.0, and v. 6.0. Available from University of Wisconsin FTP (below) at no cost. (INQUA WGDHM Newsletter 5:1-2) POLPROF þ A program for drafting pollen diagrams that produces European "Iversen-style" pollen diagrams with a modified Troels-Smith sediment column. It runs on IBM-compatible computers and drives various dot matrix and laser printers and pen plotters. A new version, running under WINDOWS, (cost not yet announced) is scheduled for release in 1993. Contact Prof. Sigmar Bortenschlager, Botanisches Institut der Universitaet Innsbruck, Sternwartestrasse 15, A-6020 Insbruck, Austria, or contact the developers at POLPROF@uibk.ac.at. (INQUA WGDHM Newsletter 6:4-6) PSIMPOLL A QuickBASIC þ program for IBM-compatible computers by Keith Bennett that generates PostScript þ page description files of pollen diagrams. Program offers user choice of diagram style, scale, font, labels, Troels-Smith sediment patterns, C-14 dates, etc. Available at no cost zipped as PSIMPOLZ.EXE from University of Wisconsin FTP (below). (INQUA WGDHM Newsletter 7:6-7; 8:11-12) TILIA and TILIAùGRAPH þ Tilia is a pollen spreadsheet program for data entry and analysis on IBM-compatible computers. The pollen data are stored in TILIA's binary format or ASCII format, but several formats can be read. TILIAùGRAPH is a powerful diagram-drafting program running on IBM-compatible computers with graphics monitors and hard disks. It can produce several kinds of curves, can plot dendrograms produced by TILIA, and drives a variety of plotters and laser printers. TILIAùGRAPH achieves this flexibility by using a "Graphical Kernel System," written by Graphic Software Systems Inc., Beaverton, Oregon. TILIAùGRAPH sells for $200.00, TILIA for $5.00 from Dr. Eric C. Grimm, Illinois State Museum, Research and Collections Center, 1920 South 10-1/2 St., Springfield, IL 62703, grimm@denrl.igis.uiuc.edu. (INQUA WGDHM Newsletter 4:5-10) (See also this issue for information on v. 2.0. Ed.) NUMERICAL AND STATISTICAL PROGRAMS CANOCO þ (ver. 3.10) A Fortran program for canonical community ordination by partial detrended canonical correspondence analysis, principal components analysis, and redun- dancy analysis by Cajo J.F. Ter Braak, Agricultural Mathematics Group, Box 1200, 6700 AC Wageningen, The Netherlands ($249.00 U.S.A.). Distributed by Dr. Richard E. Furnas, Microcomputer Power, 111 Clover Lane Dept. I6, Ithaca, NY 14850 U.S.A. CONISS þ A program, running on IBM compatibles, for clustering of stratigraphic pollen samples. available free of charge (send diskette) from Dr. Eric C. Grimm, Illinois State Museum, Research and Collections Center, 1920 10-1/2 St., Springfield, IL 62703, grimm@denrl.igis.uiuc.edu. MVSP þ (ver. 2.1) A package of common multivariate statistical procedures widely used in biology and geology, including principal components analysis (PCA), principal coordinates analysis (PCO), correspondence analysis (CA), distance or similarity measures, hierarchical cluster analysis, and diversity indices. Version 2.0 distributed as shareware by AASP Data Committee (below). Shareware edition of v. 2.1 available from University of Wisconsin FTP (below). Full- featured [*p.31 / p.32 *] v. 2.1 can be purchased for $100 U.S.A. from Dr. Warren L. Kovach, Kovach Computing Services, 85 Nant-y-Felin, Pentraeth, Anglesey, LL75 8UY, Wales, U.K., warrenk@cix.compulink.co.uk. (INQUA WGDHM Newsletter 4:1-3) (See also this issue for information on v. 2.1. Ed.) PCSLOT þ Program running on PC-compatible computers that permits sequence slotting of two pollen sequences of up to 100 samples per site. PCSLOT presents the standardized H-Matrix, allows multiple constraints for stratigraphic intervals, and computes a sensitivity analysis. Available from University of Wisconsin FTP (below) at no cost, or contact Dr. Malcolm Clark, Department of Mathematics, Monash University, Clayton, Victoria, Australia 3168, rmc@monu1.- cc.monash.edu.au. (INQUA WGDHM Newsletter 9:5-10) (See also this issue for information on v. 1.5. Ed.) POLSTA þ Program running on PC-compatibles that performs various time series analyses on stratigraphic pollen data sets. Distributed free of charge by David G. Green, Research School of Biological Sciences and Centre for Information Science Research, Australian National University, GPO Box 475, Canberra 2601, Australia, david.green@anu.edu.au. (INQUA WGDHM Newsletter 2:7-8) SLOTDEEP Program running on PC-compatible computers with graphics capabilities. Program allows manual selection of the optimum sequence on a matrix map of dissimilarities between the two sequences. Available at no cost from University of Wisconsin FTP (below). (INQUA WGDHM Newsletter 9:21-26) WACALIB þ 2.1 Program for estimating environmental variables from fossil pollen assemblages using weighted averaging and maximum likelihood regression. Distributed (for $50.00 U.S.A.) by H.J.B. Birks, Botanical Institute, University of Bergen, Allegt 41, N-5007 Bergen, Norway, birks@cc.uib.no. ZONE þ A program, running on IBM compatibles, for clustering of stratigraphic pollen samples. It is distributed, free of charge, by Steve Juggins, Environmental Change Research Centre, Department of Geography, University College London, 26 Bedford Way, London WC1H 0AP, UK, sjuggins@geog.ucl.ac.uk. Available bundled with TRAN, sample files, and manual as Tranzip.exe from University of Wisconsin FTP (below). (INQUA WGDHM Newsletter 6:4-6) FILE-CONVERSION PROGRAMS POLFILE v. 1.15 A program for converting the Wisconsin *.RAW data format to the Wisconsin *.DAT format and to a format used by Kovach's MVSP. Available as Polfil15.exe from University of Wisconsin FTP (below). TRAN þ A program to inter-convert the Tranquillini format with the TILIAùGRAPH, Cornell, Gordon, and PARADOX data formats. It is distributed, free of charge, by Juggins, Environmental Change Research Centre, Department of Geography, University College London, 26 Bedford Way, London WC1H 0AP, UK, sjuggins@geog.ucl.ac.uk. Available bundled with ZONE, sample files,and manual as Tranzip.exe from University of Wisconsin FTP (below). (INQUA WGDHM Newsletter 6:4-6) POLLEN-COUNTING PROGRAMS COUNTER þ A program that allows Macintosh and PC computers to emulate mechanical counters. Each time a key is pressed, a count is added to a category and the total is incremented. The data are stored in tab-delimited files readable by other programs. Available from Pierre A. Zippi, PAZ Graphics, 18640 Roads End Circle, Eagle River, Alaska 99577, Phone (907) 696-5702, $100. (INQUA WGDHM Newsletter 7:10-13) KOUNTER þ A program that allows PC computers to emulate mechanical counters. Each time a key is pressed, a count is added to a category and the total is incremented. The data are stored in Arizona format readable by other programs. Available from the AASP Data Committee (below). POLCOUNT A program that allows PC-compatible computers to emulate mechanical counters. Each time a 2-digit combination is entered, a count is added to a category [*p.32 / p.33*] and the total is incremented. The data are stored in Wisconsin format readable by other programs. Available from University of Wisconsin FTP (below) at no cost. (INQUA WGDHM Newsletter (8:24-27) PSION þ COUNTER A program that allows PSION þ pocket computer to emulate a mechanical counter. The data can be transferred to a PC for analysis. Available (free, send 16 KB Datapak) from Dr. K.D. Bennett, Plant Sciences, Downing Street, University of Cambridge, Cambridge CB2 3EA, U.K., kdb2@phx.cam.ac.uk. (INQUA WGDHM Newsletter 3:5) MISCELLANEOUS AASP DATA COMMITTEE PALYDISKS Distributes data files and computer software contributed by AASP (American Association of Stratigraphic Palynologists) members. Fourteen palydisks are available as of April, 1993. (IBM format, $5.00 in U.S.A. and Canada, $10.00 elsewhere). Owen K. Davis, Department of Geosciences, University of Arizona, Tucson, AZ 85721, palynolo@ccit.arizona.edu. GENERIC CADD þ Computer aided design software required by PALYPLOT (above). Available in 3 levels from $50 to $300. Generic CADD is a registered trademark of Generic Software, Inc. doing business as Autodesk Retail Products Division, 11911 North Creek Parkway South, Bothell, WA 98011-9914. UNIVERSITY OF WISCONSIN FTP Over 20 self-extracting packages of compressed files are available for downloading free of charge from "geology.wisc.edu" (or, 144.92.137.14) by anonymous FTP in the directory /pub/inqua. For more information contact Dr. Louis J. Maher, Geology and Geophysics, 1215 West Dayton St., University of Wisconsin, Madison, WI 53706 U.S.A., maher@geology.wisc.edu. E-MAIL ADDRESSES (I use the term in the general sense; it includes those on Internet, NSFNet, Janet, Bitnet, etc. A line ending with a trailing low dash [ _ ] indicates the address continues without a space on the line below.) Jim Almendinger jdinger@vz.cis.umn.edu Brigitta Ammann ammann@sgi.unibe.ch Kathy Anderson kathya@brownvm.bitnet Pat Anderson pander@u.washington.edu John Andrews andrews_jt@cubldr.colorado.edu Carlos A. Baied gg_cab@selway.umt.edu Dick Baker dick-baker@uiowa.edu Philip Barker p.a.barker@lut.ac.uk Pat Bartlein bartlein@oregon.uoregon.edu bartlein@oregon.bitnet Rick Battarbee ucfamar@ucl.ac.uk Alwynne Beaudoin userasa5@mts.ucs.ualberta.ca Pat Behling pbehling@vms.macc.wisc.edu Keith Bennett kdb2@phx.cam.ac.uk Bj”rn E. Berglund bjorn.berglund@geol.lu.se John Birks birks@cc.uib.no R. Bonnefille azerty@frmop11.bitnet Richard Bradshaw richard.bradshaw@ess.slu.se [*p.33 / p.34 *] John Brew J.Brew@vax.rhbnc.ac.uk Linda Brubaker lbru@u.washington.edu Dave Bulman dbulman@arts.cc.monash.edu.au Ian D. Campbell icampbell@nofc.forestry.ca Gail Chmura chmura@mgm.lan.mcgill.ca Malcolm Clark rmc@monu1.cc.monash.edu.au Ed Cushing cushing@vx.cis.umn.edu Les Cwynar cwynar@unb.bitnet Owen Davis palynolo@vms.ccit.arizona.edu Walter Doerfler guf12@rz.uni-kiel.dbp.de Mary E. Edwards ffmee@alaska.bitnet Scott Elias elias_s@cubldr.colorado.edu Dan Engstrom dre@umnacvx.bitnet John Flenley j.flenley@massey.ac.nz Jesse Ford fordj@ucs.orst.edu David R. Foster dfoster@lternet.washington.edu dfoster@lternet.bitnet Marie-Jos Gaillard mjgl@gemini.ldc.lu.se Konrad Gajewski gajewski@acadvm1.uottawa.ca Lisa Graumlich graumlich@ccit.arizona.edu David Green david.green@anu.edu.au Eric Grimm grimm@denr1.igis.uiuc.edu Elisabeth Gr”nlund mg@joyl.joensuu.fi Joel Guiot lbhp@frmrs11.bitnet Margret Hallsdottir mh@rhi.hi.is Sandy Harrison nguva@pax.uu.se Linda and Cal Heusser heusser@acf1.nyu.edu Sheila Hicks hicks%oygeol@figbox.funet.fi Richard A. Hodkinson umfbco5@vaxa.cc.imperial.ac.uk Geoff Hope gxh411@coombs.anu.edu.au Brian Huntley brian.huntley@durham.ac.uk Tristram C. Hussey io10651@maine.maine.edu George L. Jacobson Jr. and Heather Almquist-Jacobson jacobson@maine.edu Jan A. Janssens janss001@staff.tc.umn.edu Devra I. Jarvis djarvis@mukla.gn.apc.org [*p.34 / p.35 *] Steve Juggins sjuggins@geog.ucl.ac.uk Peter Kershaw geg625n@vaxc.cc.monash.edu.au John C. Kingston johnk@morgan.ucs.mun.edu Warren Kovach warrenk@cix.compulink.co.uk 100016.2265@compuserve.com John Kutzbach jkutzbach@vms.macc.wisc.edu Henry Lamb hfl@aberystwyth.ac.uk Gerhard Lang u346@cbebda3t.earn.bitnet Bill Last mlast@ccm.umanitoba.ca Dan Livingstone daltuc@tucc.bitnet Andre Lotter lotter@sgi.unibe.ch Ruud Lutgerink bottema@rugr86.rug.nl Glen MacDonald gmmacd@mcmaster.ca Joyce Macpherson jmacphers@kean.ucs.mun.ca Darrel Maddy d.maddy@mail.soton.ac.uk Louis J. Maher, Jr. maher@geology.wisc.edu Vera Markgraf markgraf_v@cubldr.colorado.edu John V. Matthews matthews@cc2smtp.emr.ca John H. McAndrews docjock@utcs.utoronto.ca Matt McGlone mcglonem@lan.lincoln.cri.nz Peter Minchin prm411@csc.anu.edu.au Fraser Mitchell fmitchll@vax1.tcd.ie Alan Morgan fossil@sciborg.uwaterloo.ca Robert J. Mott rjmott@emr.ca Dave Murray ffdfm@aurora.alaska.edu Dana L. Naldrett naldret@ccu.umanitoba.ca Robert E. Nelson renelson@colby.edu Bent Odgaard bo@dgu1.dgu.min.dk Jonathan Overpeck j.overpeck@omnet.nasa.gov jto@mail.ngdc.noaa.gov Marlow Pellatt mpellatt@sfu.ca Stephen C. Porter scporter@u.washington.edu Colin Prentice colin@planteco.lu.se P.J.H. Richard richard@ere.umontreal.ca Jim Ritchie jcr@aberystwyth.ac.uk UK Telephone: (0)823 42434 John Smol smolj@qucdn.queensu.ca [*p.35 / p.36 *] Tony Stevenson tony.stevenson@newcastle.ac.uk Eugene F. Stoermer eugene.f.stoermer@ub.cc.umich.edu Alayne Street-Perrott geog2@oxford.vax1.ac.uk Rachel R. Summers s.r.summers@massey.ac.nz Robert Thompson rthompso@usgsresv.bitnet W. J. Treloar w.j.treloar@massey.ac.nz Guus van der Geer gvdgeer@postoffice.utas.edu.au Adam Walanus polslask@plwrtu11.bitnet Ian R. Walker iwalker@admin.okanagan.bc.ca Jerome Ward wardjer@ducvax.auburn.edu Robert S. Webb rsw@mail.ngdc.noaa.gov Tom Webb thompson_webb_iii@brown.edu Mina Weinstein-Evron mrect36@haifauvm.bitnet Marge Winkler mwinkler@vms.macc.wisc.edu Sergei B. Yazvenko bwarner@watdcs.uwaterloo.ca Zicheng Yu yuzi@gpu.utcs.utoronto.ca Mingming Zhou zhouming@acf9.nyu.edu If you wish to be added to the directory, or to correct your present entry, please e-mail to: maher@geology.wisc.edu (If you do not receive my acknowledgment within a few days, best use regular mail as a backup.)