Feb. 2 1, I t^S4 I 



NA TURK 



Mr. Morgan's physiological definitions of reflex action, instinct, 

 and intelligence. If we want such a definition it must be made 

 independently of any zoological classification, and with exclusive 

 refennre to the foinl whether the adaptive action requires for its 

 ptrformaiice the operation of the higher nerve-centres — a point 

 which can only be determined by vivisectional experiment. In 

 other words, on the side of objective psychology the only 

 distinction that can be drawn between a reflex and an instinctive 

 action, is as to whether the action can be performed by tlie lower 

 nerve-centres alone, or requires likewise tlie cooperation of the 

 higher nerve-centre-. And this is just what we should expect to 

 find to be the case on the objective side if, a-; I have endea- 

 voured to show, the one peculiarity which distinguishes actions 

 classed as reflex from actions clasFcd as instinctive, consists in the 

 latter exhibiting in their performance a mental or conscious 

 element which is not exhiiiited i'l the former. 



Now, if the raison d'etre of the term " instinct " is thus to 

 denominate a class of adaptive actions in which there is a 

 subjective, or rather let us say an ejective element, I cannot 

 ^ee that anything but confusion is to be gained by forcing this 

 t^rm into objective implications. Were any term needed to 

 designate the neurosis of instinctive action, it would be far 

 better to coin a new one than thus to abuse an old one. I am 

 fully sensible of the difficulty which often arises in deciding 

 whether a particular action should be assigned to the instinctive 

 or to the reflex class ; but, as I observe in "Mental EvoUitio.i 

 in Animals," "this difficulty does not affect the validity of the 

 classification any more, for instance, than the difficulty of decid- 

 ing whether Limulus should be classified with the crabs or with 

 the scorpions afl'ects the vjlidityof the classification which marks 

 off the group Crustacea from the group Arachnida." 



For the rest, Mr. Morgan's criticism on my psychological 

 definition of in tinct hang, entirely upon his previous criticism 

 as to the possibility of a science of comparative psychology, and 

 as I have already endeavoured to answer the latter, 1 need not 

 go over the same ground agai i by answering the former. There 

 are only two points raised by his paper to which this general 

 answer does not apply, and with these, therefore, I shall 

 conclude. 



The first of these two points is a charge of inconsistency. My 

 critic observes that, after having said " it is enough to point to 

 the variable or incalculable character of mental adjustments as 

 distinguished from the constant and foreseeable character of reflex 

 adjustments," I go on to define instinctive actions as mental 

 adjustments whiclr are nevertheless of a constant and foreseeable 

 character. Now I think, if any one will read my chapter on 

 "The Criterion of Mind," he will see that this apparent incon- 

 sistency is not a real one. It would be a real one if the passage 

 above quoted referred only to this and that particular action of 

 an animal, apart from all the other actions of tlie same animal, 

 which, according to my criterion of mind, are competent to 

 inform us whether or not the animal in question is a choosing and 

 /e7'd-(-/r-/;/_o- animal. But the passage quoted refers to the whole 

 constitution of an animal so far as we can know it by obsei-va- 

 tion of activities, and therefore the question whether this or that 

 particular activity is to be regarded as mental or non-mental 

 (instinctive or reflex) requires to be answered by all that we learn 

 concerning the other activities of that animal. If none of its 

 activities are other than those of a constant and foreseeable cha- 

 racter, we have no reason to suppose that it is a choosing or per- 

 ceiving animal ; but if some of its other activities are indicative 

 of choice and perception, our knowledge of this fact must be 

 allowed due weight in any attempt that we may make at classi- 

 fying this or that paiticular action as reflex or instinctive. The 

 case, in short, is just the converse of that which I thus state in 

 the chapter referred to : — " Many adjustive actions which we re- 

 cognise as mental are, nevertheless, seen beforehand to be, under 

 the given circumstances, inevitable ; but analysis would show 

 that this is only the case when we have in viei\ agents whom we 

 already, or from independent evidence, regard as mental." 



The second point to which I have referred as the only one that 

 now remains for me to consider, is to the effect that I have mis- 

 taken "Mr. Spencer's position with regard to the 'very subordinate 

 importance of natural selection as an evolving source of instinct,' 

 and with regard to the question of 'lapsed intelligence.'" Here I 

 can afford to be brief, ina.smuch as any one who cares to do so can 

 compare my interpretation of Mr. Spencer's writings with the 

 passages in those writings to which I refer. It seems to me 

 perfectly clear that, although both the principles in question are 

 alluded to by Mr. ."Spencer, neither of them holds the same pro- 



minence in his theory of the development of instincts from reflex 

 action as they hold in the theory of Mr. Darwin. 



In conclusion, I trust Mr. Morgan may feel that, in writing this 

 somewhat elaborate reply to his criticism, I am marking as em- 

 phatically as I can my sense of its ability. And if the general 

 effect of this discussion is to show that the phenomena of 

 instinct jiresent peculiar difficulties to any attempt at a funda- 

 mental analysis, I should like no less emphatically to express 

 my conviction that such an analysis is not to be facilitated by 

 closing our eyes upon the entire class of phenomena to which 

 alone the word is applicable. We may, of course, abstain from 

 any attempt at such analysis, and devote our attention exclusively 

 to the physical as distinguished from the mental side of the 

 subject. Only in this case we may not speak of instinct. 



George J. Romanes 



"Mental Evolution in Animals" 



Mr. Romanes' comment on my communication in Nature 

 of February 7 (p. 335) is not quite satisfactory. I do not sup- 

 pose that he has any spite against my skate ; but as he does 

 not know me, and did not see the incident in the Manchester 

 Aquarium, I think it is very possible that he may have been 

 naturally predisposed to underrate the significance of the story. 

 I do not admit that I can be reasonably blamed for saying that 

 a repetition of the conditions would have been useful, if pos- 

 sible, while at the same time pointing out that the result would 

 mt necessarily have settled the question. Test experiments are 

 always useful, even if they do not settle the main question. 

 Mr. Romanes' terrier story was not necessary to make clear 

 what he means by "accident," and there is no analogy between 

 it and my skate story. In one case a trained, or at least tamed, 

 dog did as he was told, and the conditions of success were 

 prearranged ; in the other, a fish spontaneously did something 

 for his own advantage. .'\s for the fish smelling the food, this 

 does not harmonise with the circumstances as I described them, 

 and had Mr. Romanes seen the incident I do not think this 

 explanation would have occurred to him ; the whole series of 

 actions was too rapid, and had too much the appearance of 

 co-ordination. The propulsion of the food into the ready mouth 

 was the work of an instant. Ilsd the mouth not been ready, as 

 the cricketer's bat is the instant the ball leaves the bowler's 

 hand, the morsel would have been missed. Finally, Mr. Romanes 

 tells us (" Animal Intelligence," p. 351) that the bear observed 

 by Mr. Hutchinson was a Polar bear. Now this species is 

 "almost marine in its habits." It lives upon seal-flesh and also 

 upon dead meat which it finds floating in the water. It is not 

 infrequently cast adrift on an ice-floe or an iceberg. It is there- 

 fore not at at all improbable that the method of fishing described 

 may be an instinct developed hereditarily. The fact that two 

 bears behaved in precisely the same manner strengthens this 

 supposition. Mr. Darwin does not say whether the bear ob- 

 served by Mr. Westropp in Vienna was a Polar bear or not, but 

 he observes that the action in question " can hardly be attributed 

 to instinct or inherited habit," as it would be "of little use tn 

 such an animal in a state of nature." It seems to me that such 

 action would be very usefiil to Polar bears in a state of nature. 



Manchester, February 11 F. J. Faraday 



The Remarkable Sunsets 



At (he present stage of the discussion upon the "green sun" 

 and rosy sunsets it seems to me it would be well to recall atten- 

 tion to a few facts, for there seems to be a tendency on the part 

 of some correspondents to allow imagination to carry them 

 beyond the region of fact into that of fancy. First, then, I 

 would point out that my observations show conclusively that at 

 the time of the green sun there was an altogether abnormal 

 amount of moisture in the upper regions of the atmosphere, 

 while the ordinary hygrometric observations showed the air ne.ar 

 the ground to be comparatively dry. I have studied the rain-band 

 spectrum almost daily for the last six or seven years, and I have 

 never before known such a long continuance of the heavy rain- 

 band in a comparatively clear sky — a sky in which there was 

 only a light haze. At sunset and sunrise the intensity of the 

 bands was such as I have before seen only from an altitude of 

 some six or seven thousand feet, and even then rarely. In this 

 connection it may be well to point out that the spectrum as 

 observed by Mr. Donnelly (Nature, vol. xxix. p. 132), though, 

 as remarked by Mr. Lockyer, resembling that observed here in 



