NA rURE 



\_Fcb. 7, 1884 



lionic friction become more and more marked, and the 

 insufficiency is met by the activity of the higher centres 

 in "focusing many and more or less varied stimuli," 

 which function involves a higher manifestation of the 

 aptitude of discrimination, and as a consequence of this 

 a psychical accompaniment or consciousness. 



The author now proceeds to sketch out his general 

 scheme of mental evolution by the aid of a somewhat 

 elaborate diagram. By this last, ■w\nc\v is of a tree-like 

 form, we see how out of excitability, the distinguishing 

 property of living matter, there arises, by a double root, 

 contractility, the property of nerve-fibres, and discrimina- 

 tion, the property of ner\'e-cells, first reflex action, then 

 conscious or voluntary. In branch-like appendages of 

 the stem are represented the successive grades of 

 intellect on the one side, and emotion on the other. 

 To this are added at the sides two finely graduated scales 

 giving the products of emotional and intellectual develop- 

 ment. Opposite the numbered divisions of these scales 

 appear the names of those classes of animals, species or 

 larger groups, in which the particular products first 

 distinctly present themselves. Finally the corresponding 

 stages of mental development of the human individual are 

 appended in a parallel scale. It is only fair to Mr. 

 Romanes to say that in thus seeking to mark out by 

 definite stages or levels the progress of mind in the 

 animal series, he is fully aware of the impossibility of 

 assigning hard and fast lines of demarcation. His psy- 

 chological knowledge tells him that the several faculties, 

 sensation, perception, &c., are not absolutely distinct one 

 from another, but involve common psychical functions. 

 And his clear sense of the limits of our insight into the 

 mind of the lower animals keeps him from dogmatically 

 asserting that a particular faculty or product of mind is 

 not present below a certain zoological level. 



Having thus mapped out his ground, Mr. Romanes 

 goes on to investigate its several divisions in detail. 

 The order of treatment is as follows ; — (i) sensation, (2) 

 perception, (3) pleasures and pains, memory and asso- 

 ciation of ideas, (4) perception, (5) imagination, (6) 

 instinct, (7) reason, (8) animal emotions. This does not 

 seem a very good logical arrangement of the subject, or 

 one which grows naturally out of the diagram. It 

 appears, moreover, to make too much of the intellectual 

 side of the animal mind, and too little of the emotional. 

 This strikes one in the cursory treatment of pleasures 

 and pains along with memory, cSic, and in the somewhat 

 meagre review of the emotions in the final chapter. The 

 same thing is seen, too, in the elaborate discussion of 

 instinct, in which the highly interesting emotional 

 element in the phenomenon is hardly touched on. 



But it is, perhaps, ungracious, in view of the interesting 

 and valuable material with which the author here sup- 

 plies us, to complain of what he has not given us. To 

 touch on only one or two points of interest, the account 

 of the development of the several varieties of sensation 

 from their simplest rudiments is full and instructive. 

 The fundamental fact in memory, namely, retentiveness, 

 is clearly seized, and it is satisfactorily shown that 

 different grades of memory, e.g. mingling of traces of 

 past sensations with present ones, recalling of absent 

 sensations by association, precede the apparently simple 

 but really complex act of perception. 1 



The facts brought forward in proof of the existence of 

 imagination, that is the power of mentally picturing absent 

 objects, even low down in the scale of animals, are in- 

 teresting and conclusive. The presence in dogs, horses, 

 asses, &c., of what the author calls the third degree of 

 imagination, where the image is not suggested by external 

 objects present at the time, is ingeniously maintained by 

 the facts of dreams, delusions, and evidences of pro- 

 longed anticipation, e.g. of the stable by the homeward- 

 journeying horse, and recollection, e.g. of the lost master 

 or mistress by the pining dog. 



The piece de resistance in the volume is, as we might 

 expect, the discussion of the perplexing subject of 

 instinct. To this no fewer than eight chapters are 

 devoted. Here Mr. Romanes shows himself at his very 

 best. We see that he has mastered the wide range of 

 facts involved, and keeps the many varieties of the 

 phenomena steadily in view. We see, too, that he has 

 pondered long and well on his facts, reading what has 

 been said by others on the subject of his meditation. 

 Finally we recognise his thorough sobriety of judgment, 

 freedom from one-sidedness and from everything like 

 speculative extravagance. Mr. Romanes begins by 

 showing that instinct is clearly marked off from reflex 

 action, not merely by the degree of its complexity, as Mr. 

 Spencer says, but by its accompaniment of conscious- 

 ness. Then he proceeds to illustrate perfect instincts, 

 in which the actions are perfectly adapted to the 

 circumstances of life for the meeting of which the in- 

 stincts exist, and imperfect instincts, in which the 

 adjustment to the circumstances of the animal's life is 

 less perfect. 



This prepares the way for the main problem, the 

 explanation of the origin and development of instinct. 

 There have been two chief theories propounded to meet 

 the case. On the one hand, G. H. Lewes, and also with 

 him apparently Wundt and others, conceive of instinct as 

 a kind of " lapsed intelligence " analogous to the eflfect 

 of habit as operating during the development of a single 

 human life. Just as we come to do things in a mechanical 

 and semi-conscious way as the result of having frequently 

 done them with full consciousness, so actions of the lower 

 animals carried out with conscious design at first may, 

 as the result of long continuance in succeeding genera- 

 tions and the operation of the principle of heredity, 

 ultimately become instinctive. In opposition to this view, 

 a more humble origin has been assigned to the pheno- 

 menon. According to this theory, instinct does not 

 involve intelligence in any stage of the action. Its origin 

 is merhanical. The germ of instinctive action is due to 

 accidental variations which have become fixed and per- 

 fected by natural selection. With this view we may take 

 that of Mr. Herbert Spencer, that instincts grow out of 

 reflex actions when these reach a certain degree of com- 

 plexity, and only involve consciousness in their later 

 stages of development. Mr. Romanes combines these 

 different theories. He allows a certain weight to Mr. 

 Spencer's hypothesis as serving to explain the lowest 

 type of instinctive action occupying the border land 

 between reflex and instinctive actions proper, that is those 

 accompanied by consciousness. But fully developed 

 instincts can only be accounted for by the principle of 

 variation and natural selection, and by that of lapsed 



