Orolina exhibits four distinct varieties, all occurring together, and these differ not only in colour but in form, one variety being described as having the fore wings much elongated in the male, while another is much larger and has "the hind wings in the male different in shape." Of Heliconius Numata Mr. Bates says: "This species is so variable that it is difficult to find two examples exactly alike," while "it varies in structure as well as in colours. The wings are sometimes broader, sometimes narrower; and their edges are simple in some examples and festooned in others." Of another species of the same genus, H. melpomene, ten distinct varieties are described all more or less connected by intermediate forms, and four of these varieties were obtained at one locality, Serpa on the north bank of the Amazon. Ceratina Ninonia is another of these very unstable species exhibiting many local varieties which are, however, incomplete and connected by intermediate forms; while the several species of the genus Lycorea all vary to such an extent as almost to link them together, so that Mr. Bates thinks they might all fairly be considered as varieties of one species only.

Turning to the Eastern Hemisphere we have in Papilo Severus a species which exhibits a large amount of simple variation, in the presence or absence of a pale patch on the upper wings, in the brown submarginal marks on the lower wings, in the form and extent of the yellow band, and in the size of the specimens. The most extreme forms, as well as the intermediate ones, are often found in one locality and in company with each other. A small butterfly (Terias hecabe) ranges over the whole of the Indian and Malayan regions to Australia, and everywhere exhibits great variations, many of which have been described as distinct species; but a gentleman in Australia bred two of these distinct forms (T. hecabe and T. Æsiope), with several intermediates, from one batch of caterpillars found feeding together on the same plant.1 It is therefore very probable that a considerable number of supposed distinct species are only individual varieties.

Cases of variation similar to those now adduced among butterflies might be increased indefinitely, but it is as well to note that such important characters as the neuration of the 1 Proceedings of the Entomological Society of London, 1875, p. vii.

wings, on which generic and family distinctions are often established, are also subject to variation. The Rev. R. P. Murray, in 1872, laid before the Entomological Society examples of such variation in six species of butterflies, and other cases have been since described. The larvæ of butterflies and moths are also very variable, and one observer recorded in the Proceedings of the Entomological Society for 1870 no less than sixteen varieties of the caterpillar of the bedstraw hawk-moth (Deilephela galii).

Variation among Lizards.

Passing on from the lower animals to the vertebrata, we find more abundant and more definite evidence as to the extent and amount of individual variation. I will first give a case among the Reptilia from some of Mr. Darwin's unpublished MSS., which have been kindly lent me by Mr. Francis Darwin.

"M. Milne Edwards (Annales des Sci. Nat., 1 ser., tom. xvi. p. 50) has given a curious table of measurements of fourteen specimens of Lacerta muralis; and, taking the length of the head as a standard, he finds the neck, trunk, tail, front and hind legs, colour, and femoral pores, all varying wonderfully; and so it is more or less with other species. So apparently trifling a character as the scales on the head affording almost the only constant characters."

As the table of measurements above referred to would give no clear conception of the nature and amount of the variation without a laborious study and comparison of the figures, I have endeavoured to find a method of presenting the facts to the eye, so that they may be easily grasped and appreciated. In the diagram opposite, the comparative variations of the different organs of this species are given by means of variously bent lines. The head is represented by a straight line because it presented (apparently) no variation. The body is next given, the specimens being arranged in the order of their size from No. 1, the smallest, to No. 14, the largest, the actual lengths being laid down from a base line at a suitable distance below, in this case two inches below the centre, the mean length of the body of the fourteen specimens being two inches. The respective lengths of the neck, legs, and toe of

The lengths in the table are given in millimetres, which are here reduced to inches for the means.

FIG. 1.-Variations of Lacerta muralis.

each specimen are then laid down in the same manner at convenient distances apart for comparison; and we see that their variations bear no definite relation to those of the body, and not much to those of each other. With the exception of No. 5, in which all the parts agree in being large, there is a marked independence of each part, shown by the lines often curving in opposite directions; which proves that in those specimens one part is large while the other is small. The actual amount of the variation is very great, ranging from one-sixth of the mean length in the neck to considerably more than a fourth in the hind leg, and this among only fourteen examples which happen to be in a particular museum.

To prove that this is not an isolated case, Professor Milne Edwards also gives a table showing the amount of variation in the museum specimens of six common species of lizards, also taking the head as the standard, so that the comparative variation of each part to the head is given. In the accompanying diagram (Fig. 2) the variations are exhibited by means of lines of varying length. It will be understood that, however much the specimens varied in size, if they had kept the same proportions, the variation line would have been in every case reduced to a point, as in the neck of L. velox which exhibits no variation. The different proportions of the variation lines for each species may show a distinct mode of variation, or may be merely due to the small and differing number of specimens ; for it is certain that whatever amount of variation occurs among a few specimens will be greatly increased when a much larger number of specimens are examined. That the amount of variation is large, may be seen by comparing it with the actual length of the head (given below the diagram) which was used as a standard in determining the variation, but which itself seems not to have varied.1

Variation among Birds.

Coming now to the class of Birds, we find much more copious evidence of variation. This is due partly to the fact. that Ornithology has perhaps a larger body of devotees than any other branch of natural history (except entomology); to the moderate size of the majority of birds; and to the circum1 Ann. des Sci. Nat., tom. xvi. p. 50.

E