STUDY OF VARIATION IN THE RACES OF ZYGAENA FILIPENDULAE, L. 147 



myself that this would only have led to carrying on the confusion 

 existing hitherto between individual forms and races. The latter are 

 characterised by size, structure, tinge, etc., as well as by the extent of 

 pattern. The same name cannot be used in one sense for the entire 

 race, including these various characters, and in another sense concerning 

 only the pattern, or part of it, for an individual form which is found 

 in several races. Besides, when successive grades in the development 

 of a single character are to be designated, if we discard for several 

 reasons the simpler method of numbering them, the names given 

 should at least be descriptive and show the connection that exists 

 between the forms of the series. For these reasons I propose, for 

 instance, the following names for the increasing breadth of the marginal 

 band of hindwing, which is such a prominent characteristic of the 

 various races. In this, as in other characters, the female of each race 

 keeps at least one grade back as compared to the male. This is 

 another reason for having names independent of sex for individual 

 forms. The names already existing have always been meant to cover 

 both sexes from the same locality. I. tenuissimelimfoata : only the 

 fringes are black (not uncommon in males of subspecies fiUpetidulae ; 

 in subsp. stoecltatlis exists in female only, namely in that of calabra, 

 in some of siciliensis, and in piilcherriinaefannis). II. tenuiorelimbata : 

 a capillary streak (the usual male form in subspecies filipend ulae ; in 

 some females of siciliensis). III. tanuelimbata : streak bolder and 

 better defined (usually in male siciliensis and in female ochsenheimeri). 

 IV. iatelimbata : a narrow, but distinct band, with sinuous inner 

 outline at about J of the distance between the border and the end of 

 cell (the usual male ochsenheimeri and female etrusca),. V. 

 latiorelimbata : band broader, about 4 or -J of distance mentioned 

 above (more common male of etrusca, fiyrenes, as in Oberthiir's fig. 

 169-170, iliifioncheli, as in Duponchel's fig. 6 and in Oberthiir's fig. 158 

 and 161. and female of we/licaijiuis). VI. latissimelimbata : band 

 extending over J the distance to cell (in male of medico ginis, as figured 

 by Hiibner and by Boisduval, in many females of stoechavis and 

 atei rima and, virtually, also in the darkest forms of both sexes, although 

 its inner outline may not be discernible). It will be noticed that in 

 this series of grades the difference between two successive ones becomes 

 progressively greater as one proceeds from the first to the last. This 

 hiss come about naturally, because I have established them by a com- 

 parison of the races and not artificially. The extent of individual 

 variation within the limits of each grade is seen to increase and this is 

 evidently the cause of the increasing difference between- its averages. 

 The other characters of hindwing which lack a name are the spots 

 inside the cell and the internervural rays described above. They only 

 appear when the marginal band has reached grade V., they become 

 more and more constant and extensive as that band gets broader and 

 in grade VI. they are only absent in very rare exceptions, such as, 

 .curiously enough, Hiibner and Boisduval happen to have figured. 

 When there is only one spot at the further end of the anterior half of 

 cell I should name the form macula, employing the qualificative 

 ablative ( — " of the spot " or "with the spot"), to havea-siSihort a name 

 as possible to combine with others, when necessary. When there is a 

 second spot in the posterior half of cell I should call the form 

 bimacula and when they blend into a large one, over the median 



