Septbmbke 30, 1898.] 



SGIENGE. 



433 



We started out in the hope of finding some 

 light as to the approximate length, or, at 

 least, the approximate minimum of the 

 length, of time which is needed to trans- 

 form a race into a species, hoping that per- 

 haps those plants in which the develop- 

 ment of the individual was rapid might 

 show that in a comparatively short space 

 of time a race might be actually observed 

 to become fixed and be considered a species, 

 a fact which certainly could not be so well 

 ascertained by direct observation in the 

 study of the higher plants alone. You will 

 notice that, like the obliging shopkeeper, I 

 have not given you exactly what you ex- 

 pected, but have oiTered you instead some- 

 thing else perhaps just as good, if not bet- 

 ter. If I have not been able to tell you 

 that in such simple and quickly growing 

 plants as bacteria and Saccharomycetes 

 new species can be produced from old ones 

 in a comparatively short time, a considera- 

 tion of some of the peculiarities of such 

 plants has brought out the modifications 

 which have taken place in the views of a 

 good many as to specific limitations, which 

 is in part an answer to our primary ques- 

 tion. What do we mean by a species? 



It may be added that although some of 

 the species of lower plants may be trans- 

 formed in various ways by artificial cul- 

 tures, on the whole, we are quite as much 

 struck by their comparative constancy in 

 important respects as by their tendency to 

 differentiate. In Uredinacefe there is a 

 tendency to form adaptive races, which is 

 greater than was formerly supposed, but 

 whether the tendency is greater than would 

 be found in some higher plants, were they 

 studied as carefully as have been the Ure- 

 dinacete, is perhaps a question. Parasites, 

 as a rule, are more plastic and more sensi- 

 tive to changes of environment than other 

 plants, and their impressionability, if I may 

 use that word, might be expected to accent- 

 uate their power of specific transformation. 



It cannot be denied that there is a general 

 suspicion — to say knowledge would be too 

 strong — that the lower plants become spe- 

 cifically changed more easily and quickly 

 than the higher, but, although this is what 

 we should expect from their more rapid in- 

 dividual growth, I am not able to cite any 

 actual observations which can settle the 

 question, for, as you know, the school of 

 botanists which may be called the school of 

 ready transformationists have a fatal ten- 

 dency to accept unskillfully conducted or 

 otherwise faulty observations as convincing 

 proof. Others, it is to be feared, err on the 

 other side and are not sufficiently ready to 

 admit metamorphoses in different species of 

 the lower plants. Probably the truth lies 

 between the two. The metamorphoses to 

 which I now refer are, of course, in the 

 normal cycle of individual development 

 and should not be confused with the differ- 

 entiation into races and species, but of ne- 

 cessity our views as to the latter must be 

 influenced to some extent by our attitude 

 towards the former. 



If we turn to the second word of our defi- 

 nition which needed explanation, and at- 

 tempt to say what is meant by like individ- 

 uals, we find ourselves wholly at sea. Even 

 if we agree that the likeness must be mor- 

 phological and not physiological, that does 

 not help the matter at all. ISTo two individ- 

 uals are ever absolutely alike in morpholog- 

 ical characters, and the question is one of 

 comparative likeness only. Systematists 

 may agree that certain characters are more 

 important than other characters, but they 

 would never agree as to what characters are 

 important enough to be regarded as specific 

 in comparison with those which are only 

 racial. In fact, when we come to the point, 

 we find that most systematists do not in 

 practice distinguish species from races on 

 the ground that the former are practically 

 constant, whereas the latter are not, but 

 rather on the ground that they regard the 



