Materials for the Study of Variation Treated with Especial Regard to Discontinuity in the Origin of Species

(1) The whole of an eye-spot, centre and various concentric bands together, may be wanting ; conversely a whole new eye-spot having the centre and all the bands pertaining to the normal eyespot of the species may suddenly appear upon a crease normally bearing no eye -spot. Eye-spots therefore may come or disappear in their entirety. (2) If a number of specimens of some much ocellated species are taken and compared, examples will be found in which some of the normal ocelli are absent altogether. But besides these there may generally be found specimens having an ocellus in a reduced and imperfect condition. Speaking generally such reduction commonly occurs by diminution of the diameter of the whole spot ; but if any of its component parts are wanting the centre is the first to disappear, then the next innermost band, and so on. In Fig. 76 is shewn a series of specimens illustrating this fact in the case of Hi-pparchia tithonus. The eye-spot in its least form is represented by a plain black patch. In the more complete condition a white centre appears. A similar case in Morpho is shewn in Fig. 81. Here on the right side a certain eye-spot is absent altogether, while on the left side it is present in a reduced state ; the white centre and the innermost broad black band are absent, and the actual centre is of the yellow-red colour which in the normal eyespot of the species is the third colour from the centre. The spots

on the upper surface of the hind wings of the Wall (P. megcerd) are an excellent illustration of these principles of Variation. The principle hen- Mated, though generally followed, is not absolutely universal, and in other instances it occasionally happens that eveo when of very minute size an eye-spot still retains all its hands; l>ut the -tatement that the order of disappearance is from the centre outwards and not the reverse is substantially true. Some ha\e expre^.d a belief that ocelli arise by the breaking up "I hand- (if colour, but this view finds no support in the facts of Variation BO far as the simple ocelli of such forms as Murpho and til- Sat \ridaare concerned; for in its rudimentary condition a circular eye-spot is in them a circular eye-spot still.

The fact just stated, that in the reduction of a circular ocellus its central parts arc the first to disappear, recalls phenomena seen in many cases of disturbance propagated from a centre through a homogeneous medium. A whole eye-spot may come, or it may go (as seen in cases of Morpho), leaving the field of the cell plain and without a speck. The suggestion is strong that the whole series of rings may have been formed by some one central disturbance, somewhat as a series of concentric waves may be formed by the splash of a stone thrown into a pool. It is especially interesting to remember that the formation even of a number of concentric rings of different colours from an animal pigment by the even diffusion of one reagent from a centre occurs actually in Gmelin's test for bile-pigments. Bile is spread on a white plate and a drop of nitric acid yellow with nitrous add is dropped on it. As the acid diffuses itself distinct rings of yellow, red, violet, blue and green are formed concentrically round it by the progressive oxidation of the bile-pigment.

If the experiment is made by letting a drop of the acid fall on a piece of blotting-paper wetted with bile, a fairly permanent imitation of an ocellar mark can be made. It will be noticed that as in the natural eye-spot, so here, the outermost zone appears first and the central colour last. As also is usually the case in the ocellus, when all the zones are formed, the centre may greatly increase in diameter without any increase in the breadths of the circular zones, which merely get larger in diameter, remaining of the same breadth.

There is of course no reason whatever for supposing that ocelli are actually formed by the oxidation or other simple chemical change of the pigments of the field, but this example is merely given as an illustration of the possibility that a series of discontinuous chemical effects may be produced in concentric zones by a single central disturbance. Indeed, that the formation of an ocellus cannot be in reality of such simplicity is shewn by the fact that the scales of the centres of ocelli generally exhibit interference-colours (usually white or blue) and are then wholly or partially without pigment, while in not a few cases the centres of ocelli are deficient in, or destitute of, scales. It must also be remembered that occasionally the colour of one of the outer zones is repeated in an inner zone, which would scarcely be expected on the analog)' of the oxidation of bile-pigments.

(3) As in the case of Teeth at the ends of series, disappearance of a member of a close series of eye-spots commonly occurs by the loss of the spot standing at one of the ends of the series. This is easily seen in P. megcera, &c. Likewise as was found in Teeth, disappearance of such a terminal eye-spot is associated with reduction in the size of the other members of the series, and especially of those nearest to the place of the absent member. If as in Satyrus hyp&ranthus and many others, the series is broken into groups, then as in the case of heterodont dentitions containing gaps, a new member may be added on to the end of either group.

(4) The condition of the ocelli may vary similarly arid simultaneously in both anterior and posterior wings. In a series of Saturnia carpini for example I notice that the size of the ocelli varies greatly, those of a particular female specimen in the Cambridge University Museum being nearly a quarter larger than those of the specimen having the smallest ocelli ; but the size of the ocelli in the hind wings of each individual varies with that of the ocelli in the fore wings not less closely than the size of the right ocelli does with that of the left.

(5) This correlation between the wings of the two pairs is seen also in the presence or absence of ocelli as exhibited for instance in H. tithonus (Fig. 76). It is of course often very irregular, but for our purpose it is even of consequence that such correlation may occur sometimes. (6) As mentioned, ocelli are often coincident on the upper and lower surfaces. When this is so, the degree of development of the spots on the one surface is generally an accurate measure of the degree to which they are developed on the other surface. But in species having spots developed thus coincidently on the two surfaces it can be found that, in varying, an ocellus always first appears in its least condition either on one surface or on the other, and not indefinitely sometimes on one and sometimes on the other. In P. meycera, for example, ocelli of both pairs of wings can be seen on the under surface when not formed on the upper and conversely. Nevertheless there is always a close correlation between the degrees of development on the two surfaces.

(7) Lastly, attention is called to the circumstance that in two cases of great variation in ocellar markings there was a variation in the neuration. In the first case, P. megcera, No. 458, the second median nervure was absent from both fore and hind wings. In the fore wing upon the line where it should be there was an eye-spot: in the hind wing the eye-spots of the two cells which should be separated by the second median were partially coalescent. In the other case, S. carpini, No. 459, the large ocellus was absent from each wing, and it is stated that a nervure was also absent, but of this case no proper description has appeared, and it is uncertain which nervure was absent. When however these facts are considered in connexion with the circumstance that ocelli stand on the creases of the wings it seems likely that in some way unknown the positions and perhaps even the existence of the eye-spots may

be determined by tinmanner of stretching of the wing-membranes. It must still 1).- remembered that in the great majority of cases of ilar variation there U n<> change in the neuration. A- bo ill-- film-tin, ..f.H-ellar markings nothing is known, aiul I am net aware that any suggestion has been made which calls for The following are chosen to illustrate the general variability of ..ci-lli in Satyrida-. Any of the common forms, such as C. tlamx, P. megcera, &c. shew similar variations. Generally speaking the condition is bilaterally symmetrical, but somewhat asymmetrical examples ai-e not rare. *44.S. Hipparchia tithonus : from some 80 specimens taken in one

I-'i'.. TO. IIi/>inir,-!iiii tithoniiKi , cases illustrating Variation in number of ocelli. I. In f. \v. the upper hnlf of the large ocellus lias a pupil, the lower has none: in h. \v. no ocellus. II. I'.otli halves of large ocellus of f. w. have pupils, and the h. \v. haocellus. III. Pupils of large ocellus of f. w. are larger: h. w. has two ocelli. IV. F. w. ha- a new ocellus ami the large double ocellus is half-joined jecon 1 new ocellus. H. w. has two ocelli, one being placed otherwise than in III. V. K. w. has two ocelli without pupils as \\cll a- the large double one. H. w. hithree ocelli. The wings of the other side corresponded nearly though not riecuiately. II. is the most frequent form.

iThis figure was drawn with especial care fioni the specimens by Mr Edwin Wilson.) ditch in the Cambridgeshire Fens on the same day the individuals shewn in Fig. 76 were selected. These cases especially illustrate the statements numbered (2) and (o), viz. the order of appearance of the colours and the similar Variation of the two pairs of wings. *444. Satyrus hyperanthus : four specimens (Fig. 77) shewing FIG. 77. Satyrus hyperanthus. Various conditions of ocelli. II. is the most frequent form.

different conditions of ocelli in this species from NEWMAN'S British Butterflies. A form without ocelli is mentioned by PORRITT, Ent, xvi., 1883, p. 188. On one day I have myself taken all the forms shewn in Fig. 77 (except III.) and others in Monk's Wood, so that here no question of seasonal or local difference is necessarily involved. 44,5 Chionobas. The North- American species of this genus [in general appearance ' somewhat resembling the British Hipparchia semele, the Grayling] are of a brown colour having eye-spots on some or all of the wings. According to STRECKEK the number of eye-spots varies extremely, and the following instances are given. The species nor ma may have two spots on fore wings and none on hind wings ; two on f. w. and one on h. w. ; one on f. w. and one on h. w. ; one on f. w. and none on h. w.; three on f. w. and two on h. w. Of the species uhleri one of the types has three on f. w. and four on h. w., the other has four on f. w. and five on h. w., the subapical being very small ; other examples have only one on f. w. and two or three on h. w. The species cliryxus may have one on f. w. and none on h. w.; or two on f. w. and one on h. w. STRECKEB, Cat. Macrolepid., p. 155.

440. Arge pherusa: a butterfly resembling the British Arge galathea, the Marbled White, has a variety plesaura, in which the eye-spots of hind wing are wanting. Specimen figured in which the left hind wing is a third smaller than the right and lacks the eye-spots. FAILLA-TEDALDI, Nat. Sicil., i. p. 208, PI. xi. lig. 8. A number of species of this genus, for example, Af. <i<-l iiH-nfldiis, octavia, montezuma, &.c. are marked upon the under surface of l)oth pairs of wings with large ocelli having tour principal zones in addition to the white central spot. Of the zones the outermost i- silvery, the next dark brown, the next either red or some .-hade of yellow. Within this is a band of very variable width having a deep chocolate colour. When very broad, as in .!/. iiini,ti'ziiiii<i or M. nrhil /,'.-,-, the inner parts of this band are irregularly sprinkled with red scales. The centre is white or bluishwhite, some of the scales in its periphery being nearly ahvays di-tinct)y blue. The centre is commonly not circular but is produeed (especially in larger ocelli) in a direction at right angles to the crease on which it stands. Fig. 78, I, taken from a normal specimen ,,( .17. nrltilles, shews the usual positions of the eye-spots in all the species who-e variations are described below. The ocelli mi the tore wing are :>, on the hind wing 4. In speaking of them t he letters a, l>, c, d, e,f, g are used as shewn in the figure. Between n ami // there is a cell normally bearing no ocellus, and between d and e there are two such cells. The spot cj as described on p. 90, stands at the anterior side of its cell and not in the middle of it, and a second spot gi may appear behind it in the same cell.

The following examples are taken from the series in the collec- 1 1 rom -p. . ,1,1, us in the collection of Messrs Godman and Salviii.) tion of Mr F. D. Godman and Mr O. Salvin, to whom I am much indebted for permission to examine the specimens1. 447. Morpho achilles </. Specimen having the spots a and c entirely absent (Fig. 78, II) and the spot g very small. This specimen occurred together with two normals from Para. Ten other normal males seen, and also a specimen in Camb. Univ. Mus. having no c, the spot « being also greatly reduced.

448. M. montezuma </ : 15 specimens have all the spots from a to g of fair size. One specimen has a spot in the place a,, as shewn FIG. 79. Morplio montezuma. Abnormal specimen having an ocellus on both sides in the position al (where an ocellus normally exists in M. sulkowskii). (From a specimen in the collection of Messrs Godman and Salvin.) in Fig. 79. One specimen has a very faint a1 and gl ; another has a1 as a small ocellus, and gl indicated as a bulging of the spot rj.

In Camb. Univ. Mus. are 4 normal males and one specimen having both a1 and gl marked somewhat as shewn in the case of the abnormal M. octavia (Fig. 80). 449. M. octavia. Mr Salvin tells me that this form has a very restricted distribution and is probably only a local form of M. montezuma. In addition to 12 normal males the following were seen, all being male. Specimen having gl as a spot of moderate size ; another having g1 very small. In another a1 and gl were both present as shewn in Fig. 80. Besides these is one having g very small. All are from the Pacific slope of Guatemala. The specimen figured is from El Reposo in this district, one of the normals being from the same place.

1 In each of the figures the faint lines round the ocelli should be shewn as in Fig. 81 ; they are omitted for simplicity. ;. so. Mnrpho octavia: abnormal specimen having ocelli on both sides in the positions «' and <?, (where ocelli normally are in M. sulkowskii). (From a specimen in the collection of Messrs Godman and Salviii.) *4,50 M. menelaus £ : ten normals, and two having no a ; one having li-it a absent and right a very faint, c and y both absent. In Addition to these, the specimen shewn in Fig. 81, having no c

Fiii. Si. Mm-jilin iiii'iifliiit.-i : abnormal specimen having no ocellus c iu rt. f. w. In left f. \v. then- i.- a -mall ocellus c, luit it iriintx the tint innermost colour* of </ iioi-iiKil uci'lliix. C'lMiipared with a normal ocellus, as that at b of the same wing, the abnormal lias only \\\> z ines 1, '2 and :5, the latter colour forming the centre. 1-V. 7>, 1. may lie taken as approximately she\vin,' the normal for this species (From a -pecimen in the collection of Messrs Godman and Salvin.)

on right side, while on the left the same spot is reduced as shewn in the figure, the centre being of the colour normally constituting the third band. In connexion with the above cases it should be mentioned that in another species, Morpho sulkowskii, one of the more transparent species, the spots a1, «,, and gt are all normally present. The spot c is however sometimes absent in this species. In M. psyche the spot c is normally absent, though present in one specimen examined.

Besides the simpler ocelli there are other forms of ocelli of more complex structure, having two or more centres around which the coloured zones are disposed without an accurate symmetry. Such ocelli may be seen in Vanessa io or in Junonia, and it is noticeable that they are no less variable than the simpler forms. The following examples may be given. Vanessa io. Looking at the eye-spot on the fore wing of the Peacock-butterfly one can readily see that it is not a structure of the same nature as the other ocelli that have been already considered. The eye-spot of the hind wing does not materially differ from other eye-spots, being essentially a black spot surrounded by a pale band and containing an irregular and incomplete centre of blue. The eye of the fore wing on the contrary is not actually made up of concentric markings but is quite exceptional, being formed of a combination of patches of different colours. But whether the eye of the fore wing is a true ocellus or not it is nevertheless certain that its formation may vary with that of the eye of the hind wing, as the following examples testify.

451. Specimen, British; reared from a larva in captivity, having all the eye-spots deficient (Fig. 82). On the fore wings the series of white spots along the margin (on the creases) are present. The three which lie within the field of the normal eye-spot are lonyer than usual. The costal black mark is extended so as to cover the greater part of the FIG. 82. Vanessa io, the Peacock butterfly, having all the four eye-spots deficient (No. 4.31). (From Newman.)

situation of the eye-spot. On the hind wings the eye-spots are entirely obliterated ainl their place i- Taken by an ill-defined patch of pale colour. X r:\vM.\\, Kut., 1S72. 1 1. ln-5, Fig. 45;} Similar -pecimen in Lord \V.u.six«HAM's collection in Brit. Mus. Here the blue and black of tineye spots of the hind wing are altogether absent. The black internal border of the spot is broader than usual, and the place of the spot is lightish in colour. In the spot of the for-- \\ing the blue is deficient, the yellow is largely absent, but the white spots are emphasized.

*454. Specimen in which the eye-spots on the hind wings are obliterated, as in the foregoing : those of the fore wings are also similarly modified. I. ut the white spots of the marginal series are enlarged to a much -n-ati-r extent. Also another specimen in which the eye-spots were partially deficient These two specimens were from one brood reared in (lei-many: of this brood none were typical, and several resembled the specimens described. SOUTH, 1\., Ent., lss<), xxn. p. 218, PI.

4.")j Specimen figured in which the eye-spots are symmetrically absent from both posterior wings. In this case both the greyish yellow bordering of the eye-spots and the blue marks generally contained within them are entirely absent. The ground-colour of the hind wings i- greyish Itrosvn, and upon this two black marks are placed in tinsituation of the normal eye-spot and a series of small black lines occurs round the margins of the hind wings. The eye-spots of the anterior \\ings are modified in a peculiar manner which is not easily described. .MuSLEY, S. L., Varieties of II fit. L< i>id., Pt. ill. PI. 2, Fig. 3.

456. Junonia clelia. ('nun. In this species there are normally two ocelli m each fore wing and a similar pair in each hind wing (TitiMKN, S. Afr. JlnttertUex. i. p. '214). In a series of nine specimens in the Cambridge University Museum very great variations in the size of the ocelli appeal'. The jmxtt'rior ocellus of each wing is more constant in si/.r than the uiiti'rior. (hie specimen wants altogether the anterior ocellus of the hind wings, which in most specimens lias a diameter of about 'J-'i nun. In several the anterior ocellus of the fore wings is hardly visible.

457- Junonia cocnia : the degree to which the two eye-spots of each wing are developed vaiies greatly. In a California)! specimen in (iodman and Salvin's collection the spots are all very large, while in a (uanada specimen they are almost entirely obliterated. Of four spreimens in the same collection from the United States of Colombia (luit nut t'luin thr same locality i. one baa scarcely a trace of the anterior eye-spot of the fore wing, the si ri.iul eye 1>< ing MTV faint. In the hind wing the anterior eye-spot is very faint and th. p <sterior is absent.

The two following cases arc important from the fact that in each of them there is said to have been abnormality in neuration. which tlic second ncrvtuv d' the median vein is wanting in eacli <>f the four 'ti'iiii/ft. In the anterior wings the place which should be crossed by this nervinv is occupied by an extra ocellus (Fig. N-'!). which is nearly as large as the normal large ocellus of the wing. Tinnormal ocellus itself is incompletely doubled. In the hind wings, the two ocelli (2nd and :>rd). which in the normal insect are scpaiated by the missing nervuiv. are elongated towards each

other, so that their black borders touch and the usual central white dots join into a line, one-twelfth of an inch long. On the under FIG. 83. Pararge megara, the Wall ; case described in No. 4oS. [This copy is rather too light, and the banding on the hind wing is too distinct.] side, the anterior wings have respectively six and five ocelli and the hind wings five and six. The arrangement of the dark colour on the upper surface of the anterior wing differs somewhat in the direction of the pattern of the female. WEBB, S., Entomologist, 1889, xxn. p. 289, Fig.

Saturnia carpini ^ ; variety without eye-spots. (Fig. 84.) This specimen was bred from a larva found with many others FIG. 84. Saturnia carpini lacking the ocellar marks in each wing (No. 459). feeding upon sallow in Sawston Fen, Cambridgeshire. " In the colour and markings of the specimen there was perhaps nothing worth notice excepting the absence of the ocellus in each wing and also of one of the veins in each of the anterior wings." About 50 Iarva3 were collected at the same time on one large sallow. One of them, a female, was destitute of scales1, but the remainder of the specimens reared were remarkably fine. BOND, F., Entomologist, x., 1877, p. 1, fig. [This is the specimen mentioned by HUMPHREYS, Brit. Moths, p. 20. It is unfortunate that no further description is given, and the figure is not sufficiently clear to enable one to see which nervure was absent. On the fore wings a narrow, elongated patch of light colour was in the place of each ocellus, and on the hind wings there was a somewhat wider