January 14, 1892] 



NATURE 



249 



forest (broken up as it is into bars of light and shade), by the 

 eyes of their fierce and hungry foes as well. 



A careful examination of the varied stripes of the zebra has 

 forced upon my mind the conviction that they have a still deeper 

 meaning and value than has hitherto been noticed and 

 explained. 



It is easy to see how the vertical bars may assimilate to the 

 falling shadows in the noonday sun, and the diagonal stripes on 

 the neck and hind-quarters to those cast by the declining day. 

 But it is not so much in the day-time and during its waking hours 

 that the zebra stands in such pressing need of concealment as 

 at night, when it is compelled to rest. Then, when surrounded 

 by eager and wakeful foes, it djes require all the concealment it 

 can get. Now, let us suppose the animal to be lying down, 

 say partly on its side and partly on its belly, as horses very 

 frequently do. What will be the effect of such an attitude upon 

 the different stripes on various parts of the body? In the first 

 place, the animal will thrust out its knees, and fold its fetlocks 

 backwards under its body in such a manner that the horizontal 

 bars on the fore-legs will become vertical. 



At the same time it will push out in a backward direction, its 

 haunches, and the hind-feet will be brought forward under or 

 near its body, so that the diagonal stripes on the hind-quarters 

 will be drawn so as to become much more vertical, and to 

 correspond with the now vertical bars across the hind-legs, made 

 vertical by the folded position of the limbs. In such an atti- 

 tude — a perfectly natural and common one — all the stripes of 

 the body will be vertical, or nearly so, especially if the zebra 

 rests its head upon the ground, or its fore-legs, so as to bring the 

 diagonal stripes of its neck into unison with all the rest. Sup- 

 posing, then, that a coincidence in the general direction of the 

 stripes is produced by such an attitude of the body during rest, 

 is it too much to assume it to be an extension and refinement of 

 those protective devices of Nature, extending to the sleeping 

 zebra the full amount of all the possible protective value of its 

 stripes just at the very time when it needs it most, so that in the 

 clear tropical moonlight, when the shadows are only a little less 

 distinct than in the day, it may be able to repose in something 

 like safety and peace ? 



But, suppose the zebra rests, not always on its belly, as sug- 

 gested, but now and then on its side, with its limbs outstretched. 

 It is plain that the vertical, diagonal, and horizontal stripes 

 would then be all more horizontal than anything else, but 

 pointing in different directions, and would then so assimilate 

 themselves with the crossed and varying directions of the shadows 

 as to have the same practical effect in hiding the sleeping animal 

 from its foes. 



Under such a supposition (by no means an impossible one), 

 it seems to me that those beautiful bars of brown and white 

 surround the dormant zebra with a protection and a defence 

 almost as secure as bars of iron or brass, leaving the foes with 

 nothing but their sense of smell to guide them to their prey. 



We have only to assume the folding up of the limbs, like the 

 folding up of a two-foot rule, until the marks on both sides cor- 

 respond, and we see at once the unification in the general direc- 

 tion of all the stripes of the body, which I cannot help believing 

 has a very considerable protective value to the zebra. 



However, if any enlightened and generous patron of science 

 would kindly present to our College (" Owens," Manchester) a 

 good stuffed specimen of a recumbent zebra in the attitude I 

 have suggested, he would help considerably to settle a nice point 

 in the matter of protective colouring, and give to the cause of 

 scientific education a very welcome and appreciable aid. 



December 21, 1891. H. W. 



The Migration of the Lemming. 



There are two questions which I should like to ask Mr. W. 

 Duppa-Crolch touching his recent letter on the above subject 

 (Nature, December 31, p. 199); and for this purpose I had 

 better begin by quoting a paragraph from my own discussion of 

 the same subject, written close upon ten years ago : — 



"Looking to Mr. Collett's large experience on the subject, 

 as well as to the intrinsically probable nature of his views, I 

 think we may most safely lend countenance to the latter. The 

 most important point of difference between Mr. Crotch and Mr. 

 Collett has reference to a question of fact. For while Mr. 

 Crotch states that the migrations are made westwards without 

 reference to the declivities of the country, Mr. Collett is 

 emphatic in saying that ' the wanderings take place in the direc- 



NO. II 59, VOL. 45] 



tion of the valleys, and therefore can branch out from the plateaux 

 in any direction.' If this is so, there is an end of Mr. Crotch's 

 theory, and the only difficulty left to explain would be why, 

 when the lemmings reach the sea, they still continue on their 

 onward course to perish in their multitudes by drowning. The 

 answer to this, however, is not far to seek. For their ordinary 

 habits are such that when in their wanderings they come upon 

 a stream or lake, they swim across it ; and therefore when they 

 come upon the coast line it is not surprising that they should 

 behave in a similar manner, and, mistaking the sea for a large 

 lake, swim persistently away from land with the view to reach- 

 ing the opposite shore, till they succumb to fatigue and the 

 waves. Therefore, pending further observations on the question 

 of fact above alluded to, I cannot feel that the migration of the 

 lemming furnishes any difficulty to the theory of evolution over 

 and above that which is furnished by the larger and more im- 

 portant case of migration in general, to the consideration of 

 which I shall now proceed" ("Mental Evolution in Animals," 

 pp. 284-85). 



Mr. Duppa-Crotch's theory thus alluded to — which consti- 

 tuted the most striking feature of his "rather lengthy paper 

 before the Linnean Society," and which, he then wrote, "led 

 me to spend two years in the Canaries and adjacent islands" — 

 is, briefly, as follows : — 



" I allude to the island or continent of Atlantis. . . . It is 

 evident that land did exist in the North Atlantic Ocean at no 

 very distant date. . . . Is it not then conceivable, and even 

 probable, that, when a great part of Europe was submerged, 

 and dry land connected Norway and Greenland, the lemmings 

 acquired the habit of migrating westwards, for the same reasons 

 which govern more familiar migrations? ... It appears to 

 me quite as likely that the impetus of migration towards this 

 continent should be retained, as that a dog should turn round 

 before lying down on a rug, merely because his ancestors found 

 it necessary thus to hollow out a couch in the long grass " (Linn. 

 Soc. Journ., vol. xiii. p. 30). 



And, in a subsequent paper {ibid., p. 157 et seq.), he 

 combats the statement of Mr. Collett, "that these migrations 

 follow the natural declivities of the country." Now, however, 

 it appears that Mr. Collett turns out to be right as to the 

 fundamental fact of the migrations not being westerly more 

 than towards any other point of the compass ; for Mr. Duppa- 

 Crotch, in his letter to you, acknowledges that, in regard to this 

 point, which he previously maintained against Mr. Collett, he 

 " was betrayed into an error by trusting to common report and 

 insufficient personal experience." Nevertheless, he still main- 

 tains that the lemmings in their migrations "do not follow the 

 water-shed." 



The questions, therefore, which I have to put are : (i) What 

 are the grounds on which Mr. Duppa-Crotch continues to differ 

 from Mr. Collett touching this minor point? ; and (2) Does he 

 still maintain his theory with regard to "the island or continent 

 cf Atlantis," since he has found himself in error upon the major 

 point? George J. Romaxes. 



Christ Church, Oxford, January 6. 



Destruction of Immature Sea Fish. 



It might be supposed that the "importance of the subject" 

 would have induced Mr. Walker, at all events, to examine Dr. 

 Fulton's observations at first hand, before criticizing them 

 (Nature, December 24, 1891, p. 176), instead of confining 

 himself to reading a review. 



It may be pointed out that Dr. Fulton's computation of the 

 number of young fish captured is intended to apply only to the 

 Solway, as indeed may be gathered from your review, and being, 

 not a matter of hearsay, as implied by Mr. Walker, but founded 

 on an average of fifteen hauls extending over nine months of the 

 year, is likely to be pretty near the mark. 



In examining Mr. Walker's own computation, we find that 

 he reckons six days' fishing to the week, instead of four, which 

 is Dr. Fulton's estimate, based on local inquiry ; and we may 

 say that, if Mr. Walker has succeeded in utilizing every working 

 day during any one year of his trawling career, he must have 

 been singularly fortunate in his weather, or must have confined 

 himself to very sheltered waters. I think it will be conceded 

 that a calculation derived from actually counting the catch is 

 more trustworthy than one derived from an observation (or was it 

 only an estimate?) of weight. If, however, " 10 cwt. of young 

 flukes, . . . notonethesize of half-a-crown," is really only the 



