May 9, 1902.] 



SCIENCE. 



725 



afterwards the flowers and the fruits, made 

 their appearance, the agreement became 

 perfect. All specimens closely resembled 

 the mother, and together they formed the 

 new species, Oenothera gigas. This species 

 therefore was at once constant, even though 

 it found its origin in but a single specimen. 

 Evolved with a sudden leap from the 

 mother species, differing from it in general 

 appearance as well as in the character of 

 its various organs, it remained unchanged. 

 It was no rough cast which selection had to 

 correct and polish before it could represent 

 a distinct form; the new type was at once 

 perfect and needed no smoothing, no cor- 

 rection. 



My other species originated in the same 

 manner, suddenly and without transitions. 

 We may therefore assume that species, 

 when growing wild, do not appear gradu- 

 ally, slowly adapting themselves to exist- 

 ing conditions, biit suddenly, entirely in- 

 dependent of their surroundings. Species 

 are not arbitrary groups, as Bailey, and 

 with him many others, believed should be 

 deduced from the theory of descent, but 

 sharply defined types, unmistakable, for 

 one who has once seen them. 



Each species is an individual, says Gillot, 

 having a birth, a lease of life, and an in- 

 evitable death. Prom the moment of birth 

 until the time of death, it remains the same. 

 Only when taking this point of view can 

 we reconcile our daily experience of the 

 constancy of species with the theory of 

 descent. This is fully confirmed by the 

 results of my experiments. 



If species originated gradually, in the 

 course of centuries, their birth could never 

 be observed. Were it so, this most inter- 

 esting phenomenon would forever remain 

 hidden from us. Happily it is not so. Each 

 species as soon as born takes its place as 

 peer in the ranks of the older species. This 

 birth may be directly observed. One can 



even collect the seeds in which the new 

 types are hidden, and one can obsei-ve the 

 first steps in the development of these 

 types. Literally the new species originates 

 at the time of the formation of the seed, 

 but it is bom only at the time of germina- 

 tion. But at this period it is not recogni- 

 zable as such; this only becomes possible 

 after the first leaves have unfolded. The 

 plant can then be photographed, and in 

 this manner we may preserve the type as 

 soon as it becomes discernible and recog- 

 nizable. In fact, one can study the birth 

 of a species as readily as that of any in- 

 dividual, be it plant or animal. 



Yet it shows one important difference. 

 It is not at all necessary that a species 

 should originate in but a single specimen 

 as we saw in the case of 0. gigas. The same 

 leap, the same mutation may occur again, 

 and actually did so in my experiments, 

 where, in fact, it seemed to be the rule. All 

 that is required is that the cultures consist 

 of some thousands instead of some hun- 

 dreds of specimens. Two things then be- 

 come apparent: First, that in each lot 

 several specimens of 0. nanella, 0. lata, 0. 

 oilonga and of certain other new species 

 appear ; secondly, that it is only a few types 

 (and no others) which make their appear- 

 ance. The number of new forms is far 

 from unlimited. On the contrary, but few 

 types make their appearance annually, and 

 this among a large number of specimens. 

 There are some that are more rare, as for 

 instance 0. gigas and a most graceful, 

 small-flowered mutation which put in an 

 appearance diiring the past year. In the 

 latter, unfortunately, the seeds did not 

 ripen, and therefore, for the present at 

 least, it has disappeared, leaving no trace, 

 with the exception of a plate, a few photo- 

 graphs and some alcoholic material. 



To give a general view of the whole 

 course of my experiments on mutation in 



