PROP. CREPIN : CONSIDERATIONa ON " SPECIES." 3G9 



been cultivated, for as M. Jordan confesses, and as we shall have occasion to 

 acknowledge below, a good and complete comparison of these species, so 

 closely related, can only be made on the living specimen. However little, 

 certain individuals of each unity may oscillate between certain narrow limits^ 

 and this is a case which may bo presented, what patience must be reqmred to 

 follow, during several generations, these exceedingly slight variations. Ecforo 

 establishing any unit, it is absolutely indispensable to be assured that the 

 slightest differences — an increase of a millimetre in the length of the pedicels, — 

 a slightly increased separation between the lobes of the petals, — a little 

 darker tint in the sepals — disappear or have no permanence ; for if certain 

 individuals of any unity whatever, have preserved one or other of these 

 slight differences, if they have consequently remained stable, — such ought to 

 be placed on one side to form a new unity. A single difference, be it but of" 

 a hair, if persistent, ought, to follow out the principle logically, to suffice for 

 constituting a distinct type, for we have not any acknowledged balance for 

 weighing the value of a persistent hair — for appreciating so extremely minuto 

 a character — which may be as important as that which strikes our eyes at 

 once. This hair, from the moment that it is found to be persistent for five, 

 ten, fifteen, or twenty years of trial, should be of great value ; it ought to be 

 considered as the visible sign of a separate existence — of a " nature distinct^ 

 created in time and space, corresponding to a distinct idea, conceived from tJie 

 beginning in the Divine mind."'' Let it not be imagined that we are jesting 

 at the expense of the new school ; we do but carry out their ideas to their 

 ultimate results — results to which they must be inevitably led, if they would 

 be logical. Has M. Jordan elucidated the genus ErojyJiila with all the 

 care he claims ^ Possibly he has. The units of this genus are distinguished 

 from one another only by very small matters ; but these little things — these 

 trifles if you mil — constitute for each type an assemblage of characters whicli 

 are only valueless to the Linnean School, who require differences on which 

 they may place their finger "without hesitation, and wliich may be recognised 

 at a distance. But why also exact good and marked differential characters 

 for the species? Wliat gives the greater value to these well-marked characters? 

 It would be difficult to say. In the actual state of science, it seems to us 

 that we ought to accord provisionally, more value to the minute characters of 

 tested new species than to the larger characters of the ancient types, since, as 

 we are assured, they have persisted for several generations, and are stable and 

 unchanged ; the others should be but generic or sectional characters. 



(7) Loc. cit, p. 10. 



