THE PRESIDENT'S ADDRESS. 61 



L. alexis is chiefly in the position of certain inconspicuous black 

 spots on the under-wing. The butterfly Anosia 2~>lexipus, which is 

 spreading over the world with such marvellous rapidity, has two 

 forms, a northern and a southern ; the difference is only in whether 

 some small spots on the upper-wing are nearly white or light-brown. 

 It is the northern form that is spreading. It is difficult to see 

 how utility can have dictated these things, or the brilliant colours 

 on the inside of many bivalve shells, or the elegant shapes of many 

 asexual radiolarians, or the abortion of the index finger in the 

 Potto (Perodicticus), one of the Lemurs. Possibly the direct action 

 of external circumstances, as opposed to their indirect action through 

 natural selection, may have more influence than we are at present 

 inclined to allow. Coster states that young oysters taken from the 

 English shores and placed in the Mediterranean at once altered 

 their style of growth, and developed diverging rays like the Medi- 

 terranean species. Pony breeds have arisen quite separately in 

 different parts of the world. Great changes take place in the 

 plumage of parrots when fed on certain fish. The black shouldered 

 Javan peacock, Pavo nigripennis is considered quite a separate 

 species, and yet similar birds were bred suddenly in a flock of the 

 ordinary peacock in England without any known reason. The 

 peculiar Porto-santo rabbit reverted in England to the common form 

 in four years. The white-silk fowl reverted to the ordinary fowl in 

 England in spite of great care. It cannot, I think, be held to 

 detract from the value and importance of the theory of natural 

 selection, if we have to admit that it is not the one only cause of 

 variation. Amongst other probable causes may be mentioned the 

 inherited effect of use lately advocated by Mr Herbert Spencer, viz., 

 that an organ constantly exercised in a particular manner may be- 

 come abnormally developed in a particular direction, and that such 

 development, although to a certain extent artificial, might be 

 inherited by the offspring. Again there is the "physiological 

 selection " suggested by Dr. Eomanes, which is that the variation 

 might occur first in the sexual organs causing partial sterility with 

 the parent form, while individuals of the variety were perfectly 

 fertile inter se. This would of course require that the variation 

 should occur in more than one individual, in which case it might 

 mark off a race which might afterwards vary in other respects. 

 May we not probably, as far as our present knowledge goes, 

 reasonably come to the following conclusions ? 



