NATURAL SELECTION 269 



instances of lack of resemblance to the fosterer's eggs) as the 

 other theory. 



(3), (4) General. — The remarkable modifications of deep- 

 sea animals and of cave animals are usually given as standard 

 examples of adaptation to particularly exacting habitats. 

 Between these two categories there is common ground. In 

 both we find a tendency for the eyes to be reduced or lost, and 

 in both a compensatory hypertrophy of other sense-organs. 

 It is as well to remember that similar modifications occur in 

 other habitats where particular factors characteristic of abyssal 

 depths and caverns prevail, e.g. on muddy bottoms in shallow 

 water (Kemp, 191 7), and under rocks and in crannies 

 (Racovitza, 1907). It is a curious fact, and one which has 

 strangely enough excited little comment, that the striking 

 development of phosphorescent organs in the abyssal fauna 

 has no parallel among cave animals. Racovitza {I.e. p. 433) 

 comments on this, and states that the only phosphorescent 

 organisms in caves are some mosses and fungi. 



The occurrence of many forms in both these categories 

 which are specially modified in relation to their exceptional 

 mode of life is very well known, and there is no need to give 

 examples. The origin of these modifications has been often 

 attributed to selection. But it is not possible to discuss their 

 origin with any hope of a satisfactory conclusion, for reasons 

 which we give at length under the two separate headings. 



(3) The deep-sea fauna. — No bionomic category of animals 

 exhibits more striking or sensational examples of adaptation to 

 a special habitat than those found at great depths in the sea. 

 When, however, we start to contrast the modifications of 

 species which live habitually in deep water with their shallow- 

 water relatives for the purpose of discovering the mode of 

 origin, we encounter very grave difficulties. To begin with, 

 the technical problems are very considerable. We know very 

 little concerning the mode of life of abyssal forms, and it is still 

 largely a matter of surmise and inference. The subject has 

 been critically reviewed by one of us (Robson, 1925, 1932a), 

 and we may note the following points : 



(a) Owing to the relative infrequency of the use of closing 

 nets, there is a serious lack of knowledge as to the 

 vertical range of abyssal animals. 



