722 Mr. J. Lewis Bonhote : iSuhsjyecies [Ibis^ 



mistaken if they think that the latter alone influence heredity. 

 In my book (' Vigour and Heredity/ p. 173 et seq.) I have 

 gone more fully into the causes that may have brought 

 about geographical races, but, put briefly, it amounts to 

 this — changes of climate and food affect the " Vigour '' (rate 

 of metabolism) of an imJividual ; externally this shows 

 itself ])rinKirily in its colour, but also in many other ways. 

 The climate acting throuoh the iieneral vio-our of the 

 individual will affect the nutritive value of the egg ie.(j. 

 underfeed a breeding bird and see if the chick is not a 

 weakling), and thus the individual will tend to produce a 

 progeny having a similar vigour to itself and also of a 

 similar coloration. 



I am not suggesting, of course, that such a change would 

 take place at once, but only in the course of many genera- 

 tions ; but none the less it does take {)lace, and the conditions 

 brought about by environment are inherited. A further 

 proof of this is that when subspecies which have originated in 

 different localities extend their range and meet in a common 

 locality {e.g., the Meadow-Starlings of North America*), 

 they yet retain their subspecific characters. On this line of 

 argument it is evident tiiat discontinuous variations can 

 never prove true subspecies, since they are not due to environ- 

 ment. Dimorphic forms m:iy in some cases be due to 

 environment ; but these should, 1 think, be given speciflc 

 rank, and, in any cast', should not be regarded as subspecies 

 and designated by trinomials, since they often occur side by 

 side (e.g., some Skuas, Herons, Fulmar, etc.). My con- 

 ception of a species as a whole is that there are a number of 

 " factors " which may or may not have a Mendelian inheri- 

 tance, but which have a separate and definite inheritance. 

 A "unit" species, say the Linnet, will contain a definite 

 number of these factors ; another nearly allied " unit '^ 

 species, say the Redpoll, will contain a very large percentage 

 of the same factors, but a few different ones, and so on. In 

 dimorphic species — e. g.. Black-eared and Black-throated 

 * See Chapmau, Bull. Amor. Mus. N. II. xiii. 1900, p. 318. 



