270 Letters, Extracts, and Notes. [Ibis, 



with eggs showing the additional heavy pigmentation of 

 normal C. natalensis (fig. 20), the bird being secured at the 

 nest. Col. Rattray's interesting " hedge-sparrow " eggs 

 laid by a Nightingale (B. O. C. Bull, xxxvi. p. 79), the 

 evidence of identity being here very perfect, constitute 

 another of these cases. They strike us strongly in species 

 in which they are of relatively rare occurrence, but what 

 seems to be the same phenomenon can be illustrated abund- 

 antly from the common forms of eggs of more frankly 

 lieteroic species. That a Mendelian relationship exists 

 between such forms seems exceedingly likely, and it is also 

 probable enough that one of the main functions of the 

 Mendelian mode of inheritance is the preservation of forms 

 that have been useful in the past and may again be useful 

 or are actually still useful for the production of advantageous 

 polymorphism. 



" The wide distribution '' (I quote here from the draft of 

 my discussion of the egg-plate) " of the twin-spot tongue in 

 Passerine nestlings warns us that colour-characters that 

 probably arose in a very far-back ancestor indeed may con- 

 tinue to be utilized by its descendants in families now widely 

 divergent ; and considerations .... suggest that so-called 

 cases of coincidence or parallel evolution may be, to a vastly 

 greater extent than at first sight seems likely, instances of 

 the phenomenon " I have re-suggested above. In other 

 words, coincidence, whether it hajjpens to subserve the 

 purpose of mimicry and to receive thereby the support of 

 selection, or whether it is purely useless as coincidence, is 

 probably in the main a matter of reversion to the same 

 point in a common ancestry, and there is probably very 

 little true mimicry — based on new and independent coinci- 

 dental variation — to be found in either the eggs or mouths 

 of birds. One or two possible exceptions can be suggested. 



Holding this view, and (p. 547) not laying "excessive 

 stress on the occurrence" of mimicry, I will have seemed to 

 have devoted undue space to the subject. But it was 

 suggested to me (by more than one person) that, most 



