194 ROLLINS 



patterns, arising from genetical sources within and between populations, set 

 the stage for the development of discontinuities upon which taxonomic systems 

 may be founded. The development of taxonomy in relation to population 

 sampling is interesting in itself. During the early historical period, a single 

 sample [specimen] of a species was deemed to be adequate for study. Only 

 gradually was variation within species recognized and appreciated. With 

 this appreciation came the realization that adequate samples are required 

 to assure a factual assessment of the genetically controlled variation within 

 a species. This point of view for taxonomy is relatively recent but is now 

 firmly established. 



Learning the numerous characteristics and organizational vagaries of hun- 

 dreds of different kinds of plants as part of his "stock-in-trade" has usually 

 been the lot of the taxonomist. In order to be accomplished at spot identifica- 

 tions, he has often acquired a broad, sometimes rather superficial, acquaint- 

 anceship with a vast array of plant groups. To aid him in performing this 

 function for those who wish to know, as well as for his own ends, the tax- 

 onomist collects and assembles samples. This has been going on for over 

 three hundred years, and the number of specimens now in herbaria through- 

 out the world is well over one hundred million. In the beginning, many speci- 

 mens were collected and treated as curios. As serious study displaced curi- 

 osity, specimens began to be assembled, not only for their novelty, but for 

 the purpose of providing the data for a critical understanding of the taxa 

 they represented. The modern herbarium possesses samples of a given flora 

 in space and, in a modest way, in time. Not only is the sample thus preserved 

 the source of the data, but it may be a voucher through which observations 

 and information about more ephemeral aspects of a plant may be solidly 

 tied into a given species. It may be remarked that specimens are no longer 

 studied only visually with a dry hand lens, as Darlington (1956) would have 

 us believe, but instead provide material for thoroughgoing microscopic studies. 

 The value of specimens as vouchers of research is perhaps most appreciated 

 by taxonomists, but the need is no less great in many aspects of cytology, 

 genetics, ecology, and physiology. The number of chromosome studies and 

 published records based on misidentified material, with no chance of rectifica- 

 tion because voucher specimens were not preserved, is next to scandalous. The 

 need for specimens, in the extensive experimental work with clones of plants 

 grown in different environments, was demonstrated most vividly by Clausen, 

 Keck, and Hiesey (1940). Some further pertinent words on the uses of speci- 

 mens are given by Marsden- Jones et al. (1931). 



If the sampling of the earth's plants was at first erratic, this was only to 

 be expected, for geographical exploration itself was erratic. But with the in- 

 creased tempo of exploration came a parallel increasing interest in the plants 

 of the world for their own sake, until now we have in herbaria a minimum 

 number of samples of perhaps more than 80 per cent of the existing species 



