July 25, 1878] 



NATURE 



331 



represent the blue ray that suffers most refraction of all the blue 

 rays, and c E A the red ray which suffers most refraction of all 

 the red rays. 



It thus appears that of blue and red rays reflected from the 

 sea at the same angle, the former may reach the eye of the 

 observer and the latter not, because, though the refraction is 

 sufficient for the blue it is not so for the red ray, and it will be 

 lost in the upper air. Consequently the blue rays will appear 

 highest, and the red lowest, the other colours occupying inter- 

 mediate positions according to their refrangibility. It is evident 

 that any of these rays may be reflected too vertically from the 

 sea, and so not be refracted to the earth again, but a consider- 

 able proportion will be thus refracted, and as has been said, 

 more vertically inclined rays of the blue than of any other colour. 



When we consider the effect of rays falling to the left of c, 

 the phenomenon becomes more complicated. The same refrac- 

 tion, dispersion, and reflection take place, but the rays after 

 reflection will mostly fall short of A, and strike the sea at various 

 angles, producing a great variety of colour. It is not necessary 

 for the effect that both the blue and the red from the same 

 pencil of light should proceed to A, although this is sho\vn for 

 the sake of simplicity in the figure. It is sufficient if we know 

 that blue rays, on account of their greater refrangibility, must of 

 necessity be the highest, and the red, on account of their least 

 refrangibility, the lowest. 



If the above suggestion as to the dispersive power of the 

 atmosphere be admitted, it is probable that the question of the 

 colour and scintillation of stars will be directly affected by it. 



Little Bromley, Manningtree, July 12 R. Abbay 



Zoological Geography — Didus and Didunculus 



Mr. Searles V. Wood will, I trust, pardon me if I again 



take exception to the terms in which (supra, p. 301) he still 



writes of Z>«/«jand Didunculus. These two birds do not belong 



to the same group of Columbce. The fact that certain authors 



may have included them under the designation of "ground- 



doves" is no proof whatever of their relationship, any more 



than it is of the relationship of either to any other birds so 



called — for instance those of the Neotropical genus Chamapelia. 



I have studied pretty carefully the osteology of many forms of 



ColumbcE with especial reference to their affinities. Pezophaps 



and Didus are of course nearly allied, though even these are not 



congeners. Didunculus is at least as distinct from them as from 



Lall other Columbce with the possible exception of Otidiphaps, 



I which last I have not had an opportunity of examining. Further- 



[more, I may remark that if Mr. Wood will but look at what has 



|been published of the habits of Didunculus he will find that it 



[is as much an arboreal as a terrestrial bird, so that the name of 



I "ground -dove" is as unhappily applied to it as is that of 



~)idunculus or its ridiculous translation, " Dodlet." 



July 22 Alfred Newton 



Autophyllogeny 



The following case of Autophyllogeny, observed in a leaf of 

 Papaya vulgaris (the well-known papaw-tree) apper.rs to me of 

 sufficijent interest to be recorded in the columns of your highly 

 interesting journal. 



The letter a designates the central part of the primary leaf, 

 corresponding to the apex of the petiole on the upper side of the 

 blade. It shows some small warty protuberances, and from 

 amidst them rises a new petiole (b), about six centimetres long 

 and one and a half millimetre thick. This new petiole bears an 

 accessory leaf of somewhat pentagonal outline (c), slightly 

 crumpled and partially concave towards the upper side (the one 

 directed downwards in the figure), as if there had been some ten- 

 dency of forming a leaf pitcher. A little onwards two boat- 

 shaped appendices are observed (d and f), the midrib or petiole 

 forming their keel. They are real leaf pitchers, though of a 

 rather uncommon form. The small lateral diagram represents 

 the shape of the transversal section through f and g. The two 

 leaves are opposed to each other by their upper sides, which are 

 of a dark green colour ; the concave parts are their under sides, 

 as is proved by their pale green colour, which is generally the 

 case in the leaves of the papaw-tree. The end of the petiole 

 bears a pointed leaf (h), slightly contracted, and with a pitcher- 

 like contortion on one side. The figure is about three-fourths 

 natural size. 



The case belongs to those mentioned by Masters ("Vegetable 

 Teratology," 355, 445) under the heads of Pleiophylly and Ena- 



tion from foliar organs. His explanation is certainly correct, as 

 there cannot be any doubt that the accessory petiole b, but for its 

 development in another plane, is a true homologon of the ribs of 

 the primary leaf, and the minute warts round its base may be 

 regarded as small or checked beginnings in this same directicn. 



The described anomaly does not appear to be rare [in Fafayet 

 vulgaris. I have observed several less-developed instances ; the 

 specimen here described was given to me by one of our students, 

 Seiior Ramon Documet. A. Ernst 



Caracas, June 16 



Microscopy — The Immersion Paraboloid 



As I am responsible for exhibiting at the Ccnversazione of the 

 Royal Society, May i, the immersion paraboloid as being 

 "designed by Dr. Edmunds," I .'hculd wi>h it to be known 

 that, since that date, my attention has been directed to evidence 

 establishing Mr. Wenham's priority to the invention. 



Before exhibiting the paraboloid at the Royal Society, I had 

 Dr. Edmunds' assurance that he felt justified in requesting me 

 to describe it as designed by himself. John Mayall, Jun. 



224, Regent Street, London, July 16 



THE GENESIS OF LIMBS "■ 



in. 



T HAVE found much resemblance between the skeleton 

 ■'■ of the ventral and the dorsal fins in Kotidanus, in 

 Chiloscy Ilium, and in Raia ; also between the anal and 

 ventral fins in Notidanns. The ventral fins of elasmo- 

 branchs generally are so different from their pectoral fins, 

 and so much more like the azygos fins than the pectorals 

 are, that they serve well to bridge over the differences 

 between the orders of fins. At the same time the value 

 of the link is enhanced by the fact that in the very 

 peculiar genera Callorhynchtn and Chitnara the ventrals 

 resemble the pectorals in a very remarkable and excep- 

 tional manner. But perhaps the most instructive ventral 

 fin is that of Polyodon, the skeleton of which consists 

 simply of a double series of simple parallel rays without 

 any attachment to a pelvic cartilage which is altogether 

 absent. 



These conditions, then, appear to obliterate the dis- 

 tinctions which are at first apparent between the skeletons 



' Continued from p. 311. 



