sept. 1896] Skinner: Study of N. American Butterflies. Ill 
on low moss lands, on which water is charged with iodine, in Cumber¬ 
land, Westmoreland and Lancashire, it is a rich fulvous brown insect, 
larger and stronger built; and when these are acted upon by hydro¬ 
chloric acid gas they assume the exact color of the hill specimens. The 
dark Annulet moth ( Gnophos ob sc fir aria ) on chalk lands is a light 
colored grey or drab insect. In carboniferous limestone districts it is 
a lead-colored insect, whilst on the New Red Sandstone formation it 
varies from a rich ochreous color where oxide of iron is present in the 
soil to a dark, almost black insect on the white sandstone parts of the 
New Red formation, thus clearly pointing to geologically caused changes 
of color. Any of these latter forms acted upon by chlorine appear as 
highly colored grays. The same remarks apply to Dianthcecia car- 
pophaga. On chalk it is light buff; on ‘‘New Red” here, darker; 
but all buff in Cambrian at Llangollen; and at Penmaenbach darker 
still, buff or ochreous brown ; and on quartoise early rock, rich dark 
cold grey-brown, as in the Isle of Man, and at the Howth, in Ireland, 
ochrey shades being rarely observable upon them ; but, acted upon by 
hydrochloric acid gas they all turn to a beautiful bright light fawn buff, 
veritable carpophaga of the chalk. 
It is to be observed, however, that some varieties we might be in¬ 
clined to attribute to certain formations may be the result of a food 
proper to the soil. Thus in the cases of the Welsh Wave Moth 
(.Acidalia contiguaria ), bred continuously on heather from moss lands, 
all specimens become varieties, fumose specimens, whilst fed on succu¬ 
lent plants they are large light colored specimens, rarely darkish, but 
never so dark as when fed on heather from the moss. “ We find sea¬ 
sonal varieties not alone alternating in ordinary years, but witness their 
production by fluctuations in annual temperatures. r l hus while many 
butterflies produce one or two annual broods, in certain years, those or¬ 
dinarily single brooded become double brooded; or those which are 
double brooded produce three annual generations.”* 
I now wish to apply some of these facts to our own Lepidoptera and 
wish to say in the beginning that want of exact localities and exact data 
on our specimens has been most pernicious and detrimental to all such 
studies. In many cases specimens are without localities or dates of cap¬ 
ture or only have a State locality. Studies of variation produced by 
geographical variation in the broad sense indicated, or the effect pro¬ 
duced by seasonal broods, are impossible without such data. I also wish 
* I am indebted to Insect Variety, by A. II. Swinton, for these facts in relation 
to European Diurnals. 
