In spite of the extremely difficult economic situation in Polish Science, the majority of Polish palynologists have access to IBM-PC-compatible computers. So we started to construct programs with the main goal of moving tables of pollen data from sheets of paper to magnetic media. To encourage the cooperation of pollen analysts, we decided that the database should be built bottom-up. The software should offer to the user immediate advantages in handling pollen data, not just the promise of benefits that might come sometime in the future through participation in a European Database.
At the moment Polish Palynological Database (POLPAL) software contains programs for the input of data, numerical and graphical data reviewing and printing, and additionally, some numerical methods for the analysis of a single profile. There also is one short relational-database kind of program which allows searching for a given taxon in multiple profiles.
Input of data may be performed at three levels. The first one is the direct entry of counts into the computer while working at the microscope. The second level is input of the sample protocols after the slide has been counted, and the third level is input of tables from sites already completed. In the input of tables (with the capacity up to 300 taxa and up to 300 samples) two simplifications are available, one for low-abundance taxa with grain numbers in the single digits, and the second for taxa with long runs of zeros. At every stage errors may be corrected.
The program for graphical data presentation shows up to 50 taxa, or sums of taxa, either as percentage or concentration diagrams. The sequence of taxa may be rearranged, and the vertical and horizontal scales may be chosen arbitrarily.
The numerical analysis program contains principal components analysis, correspondence analysis, and constrained-single-link zonation. The taxonomic list now used in the database contains about 600 taxa, the list can be corrected and supplemented. It seems to be necessary to propose to palynologists a standard list of taxon names.
The special statistical features of pollen tables are exploited not only in data entering but also in data storing. As a result, on average, only half a byte is necessary for storing one pollen count. In other words one normal-density diskette may contain up to 100 large (200 taxa, 100 samples) pollen tables (Walanus, 1989).
The POLPAL program is in operation at the three Polish laboratories: the Institute of Botany Pol. Acad. Sc., Kraków, and at Gdansk and Lublin Universities. There are about 30 pollen tables stored in the database.
The POLPAL program is free for anyone interested in it. The author of the computer code (A.W.) would be happy to cooperate with the possible user in improving the program.
Reference
Walanus A. 1989. Saving computer memory in storing tables of pollen counts, Pollen et Spores, v. XXXI/1-2, p. 161-164.
Adam Walanus
Institute of Physics
Silesian Technical University
Krzywoustego 2
PL-44-100 Gliwice, POLAND
(In a letter to the newsletter coordinator, concerning POLPAL and how interested parties might obtain copies, Dr. Adam Walanus wrote "...my university is not yet connected to the 'BITNET.' I believe it will be in the near future. If some BITNET user would be interested in POLPAL please mail it, if it is neither expensive nor troublesome." With that in mind, if persons interested in obtaining a copy of POLPAL can reach maher@geology.wisc.edu by e-mail, I will send a copy on a single high-density floppy and notify Adam Walanus that I have done so. But I urge you to deal with him directly; programmers benefit from direct contact with those using their programs.)