Back in the late 1970's I recorded pollen identifications on a paper-roll adding machine as a way of recording pollen counts. It was one of the old fashioned models with 72 keys: 8 columns of keys 1 to 9. It devoted the two right columns for cents so the largest number you could enter was 999999.99. When I saw a pine grain I would ring up a 1000.01, a spruce would get 1000.02, and a sedge 1000.13. You did not have to enter the zero columns, so it was not much work. As you might guess a spruce bladder would be recorded as 500.02. You could get a count total just by touching the subtotal key. The sum of all the taxon identity numbers was gibberish, of course, but by putting an imaginary decimal point three places left of the machine's, you would get the total of all the 1's and 0.5's. This was a low-tech way of keeping a history of the pollen types so that later I could do a statistical study to see whether the different sizes or types were randomly distributed on the slide.
After accumulating several miles of paper tape, it will suddenly occur to you that somebody is going to have to go back through it all to get the total for each taxon. At the time my research computer was a "Commodore PET," with a 16K memory (upgraded from its original 8K), and I proudly worked up a program to simulate a bank of 100 counters, any one of which could be incremented by touching a two-key code. I gave the program to anyone who would take it; I do not know whether it was used.
Keith Bennett (1990a, 1990b) devised a clever aid to pollen-counting by programming a Psion "Pocket Organiser" to record taxa by a combination of one or two letters. The Psion is battery operated, relatively inexpensive, and its data can be up-loaded to a PC. I use one and enjoy it. The only problem with it is that once a count is started, you cannot do other programming with it without saving the data first. And then it is not easy to put the partial count back in the machine to continue the count. But if you think of the Psion as a dedicated counter, this is no problem; it simply shuts down and "sleeps" when you stop touching the keys.
I never bothered to use my IBM PC as a pollen counter; it is in constant use for other things. But then I noted Pierre Zippi (1992) had a program for turning a Macintosh into a microfossil counter, and I got to thinking that a lot of labs probably have old outmoded DOS PCs around that are being used mostly as furniture. So I reworked my old PET program by borrowing some of Keith Bennett's Psion ideas and added some save and retrieve features. Owen Davis comments in this issue (p. 14) that we ought to deal with counters for microfossils, so this is a first shot. If others let me know what they have developed, I will include them in the next issue.
POLCOUNT sets up 100 counters for 100 different pollen taxa. Think of these counters as labeled 00, 01, 02, ... 99; each is assigned to a pollen taxon. Each time POLCOUNT is run, a *.TAX file is loaded that associates the name of a real object (a pollen taxon, for example) with each of the counters. You can load TAXLIST1.TAX, a dummy taxon list that comes with POLCOUNT, in order to see how the program works. To record a grain of taxon 05 in counter 05, simply type the two numbers 0 and 5 in sequence. You will hear a click, and the actual name of taxon 05 will appear on the screen as confirmation. You do not need to touch the <Enter> key; the program knows that any two strokes on number keys should increment the named counter by one. You will note that keys other than numbers usually produce an irritating beep. (The <Esc> key will always end the program if you wish to do so.)
The number keys serve to increment the counters. One of the 100 counters you control is reserved for a special purpose: 00 is used for marker grains that you might add to the sample for determining pollen concentration. There is also a counter 100, but you cannot access it directly because 99 is the largest two-digit number you can enter. Counter 100 sums the counts in counters 01 through 99 (it does not include counter 00); you can think of it as keeping a rough record of the pollen sum.
POLCOUNT loads its Taxon List each time the program is run. It offers TAXLIST1.TAX as the default name, and you can choose that by merely touching the <Enter> key. The initial TAXLIST1.TAX file is a dummy sample that allows you to see how the program works. TAXLIST0.TAX is an 'empty' list consisting only in the names: Marker Grain, Taxon 01, Taxon 02, etc. When you decide the taxon order you wish to use, load TAXLIST0.TAX into your word processor and replace the dummy names with the actual taxon names you are going to use. The names can be short or long; they can contain spaces. When you wish to look at the taxon list while running POLCOUNT (see F7, below), only the first 8 letters will show. SAVE YOUR LIST IN ASCII TEXT FORMAT under a different name, but ending with the extension .tax. If you use TAXLIST1.TAX, it will load as the default. Different analysts can use different lists simply by giving them different names (But always with the same 3-letter extension: '.TAX') which can be entered manually from the *.TAX files shown in the directory listing.
The function keys F1 - F10 allow you to correct errors, seek information, edit the counters, and load and save data files. The purpose of the function keys is always displayed at the bottom of the screen while counts are recorded:
Two POLCOUNT sample files are shown below. POLCOUNT makes an ASCII text file, fashioning a title line from the site's name and the day's date. The second line is always 102, the number of categories in the file: counters 00 through 100, plus the sample's identifying number, usually its depth in centimeters. You will note that all data lines (except the last) consist of ten values, each separated by a space. The sample's depth in centimeters is the first value on the file's third line; the other numbers in the sequence are the sums in counters 00 through 100. The sum in Counter 100 will equal the total of all the previous categories EXCEPT THE FIRST TWO, which are Depth and the number of Markers.
Blue Lake, WI = 01-02-1992 102 .5 250 1.5 0 7 12 1 1 16 5 0 33.5 13 172 5 3 5 10 5 1 1 1 1 2 0 0 2 0 0 0 0 0 0 0 0 24 11 5 153 4 13 0 1 1 4 2 0 1 2 3 0 0 0 0 0 7 5 2 21 38 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 595 Blue Lake, WI = 01-11-1992 102 535 315 30.5 1 117 0 5 23 123 8 44.5 153.5 44 59 10 7 11 4 0 0 0 0 0 1 0 1 0 0 0 0 0 0 0 0 0 6 13 5 3 2 2 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 9 5 1 40 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 729.5The following is the partial list of taxa in TAXLIST1.TAX. The file has a title line (here, BlueLake Taxa), followed by the number 101, which is the number of categories to follow.
BlueLake Taxa 101 Marker Picea Larix Fraxinus ... Taxon 98 Taxon 99 Sum(01-99)A "Wisconsin" format file has a site title for the first line. The second line records the number of taxon categories in the file, and the third line records the number of samples in the file. The sample counts then follow in sequence from top to bottom in the core, each starting with an identifier such as the sample's depth in centimeters. The 102 taxon categories are found at the end of the file, each on a separate line. Use a word processor to make a "Wisconsin" format file. Load each POLCOUNT sample in sequence into a single large file and append the TAXLIST1.TAX file at the end. Use the word processor to delete the first two lines of each sample (the title and the number 102). Put a main title at the top, followed by the number 102 on the second line and the total number of separate samples on the third line. From the appended TAXLIST1.TAX list, delete the first name (the title) and the second line (101). Insert the word Depth(cm) on a separate line just after the counts and just before "Markers." The following is an example of a "Wisconsin" format file made from the example material.
Blue Lake, Wisconsin 102 2 .5 250 1.5 0 7 12 1 1 16 5 0 33.5 13 172 5 3 5 10 5 1 1 1 1 2 0 0 2 0 0 0 0 0 0 0 0 24 11 5 153 4 13 0 1 1 4 2 0 1 2 3 0 0 0 0 0 7 5 2 21 38 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 595 535 315 30.5 1 117 0 5 23 123 8 44.5 153.5 44 59 10 7 11 4 0 0 0 0 0 1 0 1 0 0 0 0 0 0 0 0 0 6 13 5 3 2 2 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 9 5 1 40 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 729.5 Depth(cm) Marker Picea Larix Fraxinus ... Taxon 98 Taxon 99 Sum(01-99)When the "Wisconsin" format file has been created with your word processor, save it as an ASCII text file with the extension .RAW, such as BLUELAKE.RAW. This file will undoubtedly contain some taxa that never received a count in any of the samples. You can remove the taxa with zero counts from the file by using the program REMZEROS. Run REMZEROS and load "BLUELAKE.RAW." You can save the file without the unused taxa either with the same name or another of your choice. But it is good practice not to use the same name; keep the original file intact in case you wish to count additional levels later. The new levels can be put in their correct positions with a word processor and REMZEROS run again to remove any taxa still with zero counts.
If you wish free copies of POLCOUNT and REMZEROS, I would be pleased to make them available both as compiled programs and as QuickBASIC source code. The latter will work with QBASIC that comes with DOS 5.0, and will allow you to modify the programs to fit your own needs and output file structures. These programs make no use of graphics and should run on any IBM compatible PC. The pollen counter package which contains the programs, sample files, and instructions for transferring the final counts into TILIA, is available in a self-unzipping format as POLCNTPK.EXE. This is also available for Anonymous FTP in directory /pub/inqua of geology.wisc.edu (see the article on the Internet on p. 17 of this issue).
Bennett, K. D. 1990a. Pollen counting on a pocket computer. INQUA - Commission for the Study of the Holocene, Working Group on Data-Handling Methods Newsletter 3:5.
Bennett, K. D. 1990b. Pollen counting on a pocket computer. New Phytologist 114:275-280.
Zippi, P. A. 1992. Scientific software for Apply Macintosh. INQUA - Commission for the Study of the Holocene, Working Group on Data-Handling Methods Newsletter 7:10-13.