54 



NA TURE 



[May 2 1, 1896 



west, what has been pointed out concerning the fading of the 

 rest of the sky, the comparative localisation of light and colour 

 in the west, and the illusion as to brightness that occurs when 

 the background fades, the phenomenon of AlpenglUhen, and 

 that of the greenish illumination so often seen, can be accounted 

 for without the help of the startling hypothesis quoted. 



But it would be more satisfactory if observations could be 

 made from above. Would M. Vallot sacrifice himself and spend 

 some nights up in the observatories that he directs? 



R.N.E. College, Devonport. W. Larden. 



The Positions of Retinal Images. 



The thanks of your psychological readers are due to Mrs. 

 Ladd Franklin for having, in her letter published in your 

 number of February 13, called attention to Schon's experiments, 

 which, as she says, have been unaccountably overlooked. I have 

 in consequence been repeating the experiment which Mrs. 

 Franklin describes, but so far with purely negative results. 

 Although some of the observers gave answers which might 

 hastily have been interpreted as confirmations of Schon's 

 illusion, a further analysis showed conclusively that no one on 

 whom I have experimented, so far, perceived it. 



Allow me to indicate one or two points in Mrs. Franklin's 

 letter which seem to require elucidation. She writes as if the 

 object looked at in the experiment " consists of a single bright 

 point." But surely the point H in her diagram — the fixation 

 point — was a bright point as well as o or o' ? In Schon's 

 experiments the apparent distance of O or O' was judged 

 relatively to H (which was a stick of phosphorus), by the 

 *' stereoscopic (or pseudoscopic) effect," and his explanation of 

 the illusion was that we (unconsciously) judge as if the 

 image actually produced on the right retina had been produced 

 on the left, and vice versa. The image on each retina 

 consists of two bright points, but cannot strictly be 

 called a " double image," since the bright points are 

 produced by two distinct objects — by the phosphorus at H, and 

 by the light proceeding from o' or o. If Schon's explanation is 

 correct, then, supposing the light really proceeds from o', and 

 when the ray o' ;•' is darkened appears to come from o, the 

 observer ought to say the object appeared to be not merely as 

 far off as H, but a /<?«,? way behind it. Further, if the ray o' / 

 is darkened instead of o' ■/ there ought to be no illusion — he 

 should say the object appears much nearer than H (i.e. still at o') ; 

 and if either ray is cut off altogether, he will have no reason for 

 judging the object to be at o', but will probably judge it to be 

 further back — where the source of light actually is. In my 

 experiments, so far, none of the observers have made any 

 distinction between cases where the ray o' r' was darkened and 

 those where the ray o' / was ; but if either of them was darkened 

 considerably, they answered just as they did when one of them 

 was totally extinguished ; judging the object to be about where 

 the source of light actually was — which was about the same 

 distance as the phosphorus mark H, and very much nearer than 

 the point o would have been. 



I hope to continue the experiments, if possible until I get a 

 positive result, and should be glad therefore to hear some 

 further details of Mrs. Franklin's experiments, especially with 

 reference to the points I have brought forward, either privately 

 or through your columns. Edward T. Dixon. 



4 Cranmer Road, Cambridge, April 17. 



Colour Variations in Ducks and Pigeons. 



About a year ago you published a short article by Mr. Francis 

 C'lalton (April II, 1895, vol. li. p. 570), in which he urged the 

 desirability of making careful records of all cases of "sports" 



) sudden variations in domesticated animals, &c. Two such 

 sports having arisen recently under my own observation, one in 

 ducks and one in pigeons, I write to place the facts before your 

 readers. 



(l) Ducks. — In January 1894, I bought in Beyrout market a 

 drake of the common " Mallard ' colours and four ducks, two of 

 normal wild-duck colour, one pure white, and one black, splashed 

 with white. From these ducks I raised, the same season, thirty- 

 six ducklings ; and, from eggs given by a friend, nine more. 

 Concerning the latter, nothing need be said at present, except 

 that their own mother was of a very dark, dingy brown, and the 

 ducklings were nearly black in the down. Of the thirty-six 

 ducklings hatched from my own ducks' eggs, twelve or thirteen 



NO. 1386, VOL. 54] 



(I neglected to note the exact number at the time) were different 

 in colour from their olive-green brethren and from anything I had 

 seen before, being of a beautiful pale fawn colour above, shading 

 into canary-yellow beneath, with darker pencillings and shadings 

 on the sides of the head and back, and with the normal, sym- 

 metrical series of three pairs of light marks on the upper 

 surface, distributed just as in normal, olive-green ducklings. 

 The entire set of these pale ducklings proved to be females, and 

 their plumage, when adult, was a pretty yellowish or sandy buft 

 colour, with darker shadings, due to a brown streak down the 

 middle of each of the contour-feathers. The speculum on the 

 wing gave mostly sky-blue reflections, instead of the usual 

 metallic green of common ducks. Two only of the dozen (or 

 thirteen) difiered perceptibly from the others, being of uniform 

 cinnamon-brown colour, with white throats. 



Five of these pale ducks were kept and allowed to breed, viz. 

 one cinnamon-brown and four j'ellow ones. In addition, my 

 stock during the season of 1S95 consisted of three of the original 

 old ducks (one white, one black, and one normal) ; three normal- 

 coloured young ducks related to the pale ones {i.e. same pater- 

 nity, and presumably same maternity to some extent) ; and two 

 ducks raised from the eggs given by my friend, as above 

 mentioned, and therefore non-related to the others — in all 

 thirteen, ducks. Of drakes there were four — two of normal 

 mallard colour (related, as above, to the pale ducks), and two 

 own brothers to the dark ducks, these having green heads and 

 beautifully-pencilled stone-grey bodies, with no brown on the 

 breast and no white collar — a departure from typical drake- 

 colouration which is normal (in Syria at least) to dark varieties. 



From this stock of ducks I raised last spring sixty-two 

 ducklings, of which nineteen were fawn-coloured in the down. 

 One of these died very young. Of the remainder, fourteen were 

 females and four males. All were sandy-buff, none cinnamon- 

 brown ; but one— a female — was a shade or two darker than 

 the rest, and when adult showed no metallic colours on the 

 speculum, agreeing in this respect with the dark ducks of alien 

 parentage. 



Of greatest interest to me was the question : What will the 

 " yellow" drakes be like when adult? Time has answered as 

 follows : Head and neck, 'io{\. coffee brown, with obscure greenish 

 reflections in some lights ; narrow white collar ; chestnut-brown 

 breast, similar to mallard ; upper tail-coverts (including curled 

 feathers), and under-tail coverts, chocolate-brown ; the rest 

 delicate cream colour, with fine transverse pencillings on back 

 and sides, similar to those on the mallard, but paler and less 

 distinct : the whole effect very pleasing. 



Of course all this may be familiar enough to some people, 

 but it is quite new to me, and no mention of such drakes is 

 made by Darwin in '•.\nimals and Plants," nor by any other 

 writer whose works I have been able to consult. Whether 

 atavism has anything to do with the matter, I cannot say, as the 

 parentage of my original stock is entirely unknown ; but I am 

 accustomed to notice very carefully all the ducks I see about 

 town and the surrounding country, and am sure I have never 

 come across any such during an experience of about twenty-five 

 years. In any case, it is interesting to note that the new variety 

 was far from being "swamped " by the inevitable crossing with 

 its parent form. 



(2) Pigeons. — In 1894 I procured a pair of birds of a variety 

 known to Arab fanciers as black Urjani (or ShainandarAzi). 

 These are largish pigeons, wholly black, with two "red" (i.e. 

 bright reddish brown) bars on each wing, corresponding to the 

 black bars on normal "blue" pigeons. The pair were un- 

 related, the male coming from Hums, the female from Damascus. 

 The variety is scarce in Beyrout, and is valued more or less by 

 all Syrian fanciers, who breed it with some care ; and it habitually 

 breeds true. My birds produced during the season of 1895 ten 

 young ones : six (3 <J , 3 9 ) quite normal in colour ; one ( 9 ) 

 slightly mottled on the shoulders with brown .and a very little 

 white ; and three (all 9 ), which in the nest plumage were 

 uniform light red. (I had not a red bird in the loft — scarcely a 

 red feather, aside from the red bars of the Urjanis themselves, so 

 there was no question of illegitimate paternity.) But, strange 

 to rel.ate, when these red birds moulted, nine-tenths or more of 

 their red featliers were replaced hy pure white, so that their adult 

 plumage ma)- be described thus : white birds with red neck, 

 abdomen red mottled with white, a very few red feathers 

 scattered over back and shoulders ; no trace of red bars. 



Careful inquiry among Arab fanciers having personal ex- 

 perience of the breed in question, elicited the following 



