18 BRITISH BUTTERFLIES. 



elevation. have none. We believe it will be found, as we have 

 already noted, that the cause of the difference is more likely 

 to be discovered in the geological nature of the ground, coupled with a 

 denseness of the herbage, etc., or the reverse, in the locality, i.e., in 

 the character of the more exposed rocks. Beverdin remarks 

 that he has a 2 with the spots of the submedian series on the hind- 

 wings very small, whilst the ' white rings surrounding them are 

 abnormally wide, giving the specimen a very special appearance. 

 The variation in the number of spots is fairly extensive, and may 

 be in the direction of increase or decrease. The' submedian row of the 

 forewing presents rarely an 8th (apical) spot, but there is occasion- 

 ally an increase in the number of spots, -especially between the sub- 

 median and discoidal, the basal and discoidal, or just outside the sub- 

 median. These examples are known as ab. addenda. On the other hand 

 the number of spots is often considerably reduced (ab. obsoleta, Tutt), until 

 in the most extreme forms there are only the discoidal and marginal 

 lunules left = ab. corydonis, Bergstr. (cinnus, Hb.); beyond this is a form, 

 usually of the pallida or albescent underside type, in which the marginal 

 lunules and the discoidals are further reduced almost to obsolescence 

 (obsoletissima, n. ab.). It frequently happens that the spots on the hind- 

 wings are absent, and those on the forewings normal ( = ab. ciiuuis, Gerh.), 

 the form in which the spots on the forewings are absent, and those on 

 the hindwings normal ( = ab. antico- obsoleta) are rare. Forms in which 

 one or three wings are obsoletely marked and the others normally 

 spotted (irregularis, n. ab.) are not uncommon. Forms with the right 

 side obsolete and the left side normal (dextro-obsoleta, n. ab.), and others 

 with the left side obsolete and the right normal (sinistro-obsoleta, n. ab.), 

 are very rare. Pickett observes (in lift.) that, in his collection, out of 51 

 $ and 33 2 obsoleta (sens. lat.),5 $ s and 5 $ s are antico-obsoleta, 35 $ s and 

 15 $ s cinnus, Gerh., 4 $ s and 4 2 s corydonis, Bergstr., 3 £ s and 6 J s 

 dextro-obsoleta, and 4 $ s and 3 $ s sinistro- obsoleta. The addenda 

 form is sometimes further modified by the union of the normal and 

 supernumerary spots into striate form. This is particularly the case on 

 the forewings when the normal submedian spots are extended and united 

 to the extra spots to form a series of black linear streaks pointing in the 

 direction of the discoidal (usually without reaching it), the basal spots 

 (the upper particularly) forming a streak (or streaks) in the direction of 

 the discoidal, also usually without touching it (=ab. juncta), and this 

 may occur only on the fore- or hindwings (antico-juncta and postico- 

 juncta) the other wings being normally spotted. In others the direct 

 union of the submedian and basal spots with the discoidal (ab. striata, 

 Tutt) shows little trace of the intermediary of supernumerary spots, hi 

 other examples, again, only the lower basal and lower submedian spots 

 are joined ( = ab. parisiensis, Gerh.), occasionally this line is double 

 (biarcuata, n. ab.), sometimes the spots approach, without joining (se)iri- 

 arvuata, n. ab.). The union of the first basal and the first submedian 

 spots of the hindwing ( ab. costajuncta, Tutt), the third basal and the 

 seventh submedian ( — ab. basijuncta, Tutt) are also not uncommon. 

 Forms in which parisiensis, costajuncta, and basijuncta are combined 

 are rare (confluens, n. ab.). Union of striate and obsolete conditions in 

 the same specimen is not uncommon, e.g., ab. <l<>his. Ilk. ab. anticoex- 

 tensa- obsoleta, etc. (see infra). The modification of the underside 

 margin is most interesting. Normally there are seven marginal 



