NA TURE 



237 



THURSDAY, MAY 18, 1916. 



MIMICS READY-MADE. 

 Mimicry in Butterflies. By Prof. R. C. Punnett. 

 Pp. vi+i88 + xvi plates. (Cambridge: At the 

 University Press, 191 5.) Price 155. net. 



THE scope and general arrangement of this 

 work are indicated in the following list of 

 its eleven chapters : (i.) -^ short introduction on 

 teleological interpretations— theolc^ical and other- 

 wise ; (ii.) A historical account of Batesian and 

 Miillerian mimicry; (iii.) Old-world mimics, with 

 a ver}- poor reproduction on p. 19 of Dr. Eltring- 

 ham's illustrations of the fore-feet of butterflies; 

 (iv.) New-world mimics; (v.) Criticisms of "the 

 five conditions which Wallace regarded as con- 

 stant for all cases of mimetic resemblance "; (vi.) 

 "Mimicry rings," a discussion on the origin of 

 mimetic resemblances and initial steps ; (vii. and 

 viii.) On Papilio polytes— the Mendelian relation- 

 ship between its female forms and their origin ; 

 (ix.) The enemies of butterflies; (x.) Mimicry and 

 variation; (xi.) Conclusion, summed up in the last 

 words — "The facts, so far as we at present know 

 them, tell definitely against the views generally 

 held as to the part played by natural selection in 

 the process of evolution " — ^viz., against the 

 theory that adaptations are built up by the 

 gradual accumulation of small variations. 

 . The last chapter is followed by two appendices, 

 the first containing a table by Mr. H. T. J. Norton 

 giving the means for "estimating the change 

 brought about through selection with regard to a 

 given hereditary factor in a population of mixed 

 nature mating at random " ; the second explain- 

 ing the differences between the three sections of 

 Papilio, and giving a list of Papilionine models 

 and mimics quoted in the text. 



The principal feature of the book is its illustra- 

 tion by means of twelve excellent coloured and 

 four uncoloured plates. There are unfortunately 

 a good many errors and much want of judgment 

 in arrangement and in some of the examples 

 selected. 



In so complicated a subject as mimicry it is 

 a great help to the reader to adopt some uniform 

 s>'stem in the arrangement of models and mimics, 

 and for many years it has been a usual custom 

 when the figures are side by side to place the 

 mimic to the right ; when they are one abo^■e the 

 6ther, to give it the lower place. The present 

 work adopts no system at all. Sometimes, as in 

 plate vii., the mimics are to the right; sometirhes, 

 as in viii. and xv. , they are to the left; and so 

 with upper and lower. 



- There are also unfortunate errors in the naming. 

 Fig. 3 on plate i. is certainly not Danais septen- 

 trionis, but a Radena, probably R. vulgaris. The 

 former butterfly is nearly represented' by the 

 cfosely allied D. petiverana,' shown on plate vii, 

 fig. I. A still rrtore serious mistake occurs on 

 litts last plate, where the names of figs. 2 and 

 3 are transposed in the description and in the 



NO. 2429, VOL. 97] 



text, so that a Danaine model is made to bear 

 the name of the Papilionine mimic of another 

 model, and vice versd. Apart from this, the 

 model shown in fig. 2, if only one was to be 

 figured or mentioned in the text, is not well 

 chosen, and it is natural that the author should, 

 on pp. 29, 30, criticise his own selection. 

 Amauris echeria and albimaculata are the well- 

 known models of the brasidas form of Pap. leon- 

 idas in the south and south-east parts of its 

 range. The same Danaines are also deprived of 

 their true place as the models of Pap. echerioides, 

 being ousted by Am. psyttalea in the taWe on 



P- 159- 



The descriptive title of plate xii., "South 

 American Butterflies," is unfortunately chosen, 

 for the lowest of the four figures is a moth, and 

 the word " Butterflies " in conspicuous capitals 

 immediately beneath the figure quite overshadows 

 the diminutive " (Heterocera) " at the side. Plate 

 XV., "illustrating the closely parallel series of 

 patterns occurring in the two distinct groups 

 Heliconinae and Ithomiinae," is unfortunate, both 

 in the names and in one of the genera selected — 

 Mechanitis. If a single Ithomiine genus was 

 to be shown with Heliconius, it should have 

 been Melinaea, the undoubted primary models 

 of the Heliconines and almost certainly of 

 the species of Mechanitis as well. The resem- 

 blances shown on plate xv. are, in fact, the 

 secondary or incidental resemblances between 

 species that mimic the same models — not them- 

 selves illustrated. As regards the names, it is 

 perhaps too much to expect a writer whose main 

 interest is bionomic and evolutionary to follow all 

 the ups and downs of synonymy. But the ex- 

 amples are not numerous, and it is easy to get 

 assistance from friends devoted to the study of 

 systematics. Furthermore, most of the^ examples 

 on plate xv. had already been figured and named 

 in the excellent, although uncoloured, plates 

 XXX.— xxxiii. of J. C. Moulton's paper in Trans. 

 Ent. Soc., 1908. Of the five species of Heliconius 

 figured on plate xv., fig. i, minis is regarded 

 as a form of novatus ; fig. 2, telchinia, of isme- 

 nius; fig. 3, eucrate has been long known as 

 narcaea narcaea; fig. 5, " splendens," a name un- 

 known in the genus (splendida, Weym. , does not 

 resemble the figure), is aristiona hicolorata. Fig. 

 10, Mechartitis "meihona" is doubtless intended 

 to be M. decepius, the true co-mimic of the ac- 

 companying Heficonius (fig. 5); but a butterfly 

 from a different association and from farther 

 north, ^f. messenoides, has apparently been 

 figured — either this or a form transitional be- 

 tween it and decepius. "Methona " is a third 

 rendering of Hewitson's moihone. Salvin having 

 introduced a second rendering, " methone" ; b«t 

 the butterfly originally named by Hewitson is a 

 Melinaea, arid not a Mechanitis at all. 



Plate xvi. and the corresponding parts of the 

 text suffer from the oriiission of a third North 

 American Daname from Arizona, D. strigosa, 

 and the corresponding Limenitis, L. obsoleta 

 (hulsti), which, although an excellent mimic, 



N 



