272 Mr. T. Vernon Wollaston on 
took a considerable series of it from his vineyard at 8. 
Joao, near Funchal, hoping that the examples thus ob- 
tained might enable me to clear up one or two doubtful 
points concerning some of the closely-allied forms which 
I had originally published as specifically distinct, but two 
of which I afterwards suppressed,—feeling it more pro- 
bable that they were in reality but local races of the 
(evidently variable) A. noctivagans. Accordingly, having 
hkewise captured a long array of individuals, barely 
differmg at first sight from the others, at 8. Antonio da 
Serra, I have been examining the two sets with unusual 
care,—being satisfied that if there is more than a single 
species concealed amongst the three forms which I 
admitted into my emended definition of the A. noctiva- 
gans in 1857, two at any rate would be likely to present 
themselves amongst my series from regions so dissimilar, 
and remote, as the vineyards around Funchal and the 
elevated mountain-district of S. Antonio da Serra. The 
result is that, despite the prima facie resemblance of the 
whole, I cannot but believe, as I did originally in 1854, 
that, after all, there must be two species indicated (one 
found in the higher altitudes, and the other in the lower), 
and that consequently I was mistaken when, in my sub- 
sequently-published (and re-adjusted) Madeiran Cata- 
logue, I referred them both (contrary to my original 
conviction) to a single plastic type. Yet at the same 
time the extreme difficulty of ascertaining the true 
specific limits of these variable, scale-covered Cyclomides 
must be my excuse if even now I am in error, when 
endeavouring to re-instate at all events one of the two 
forms which, although treated in the ‘Insecta Made- 
rensia’ as truly specific, I afterwards suppressed. 
Since the true A. noctivagans (as enunciated by me in 
1854) clearly attains its maximum in the laurel regions 
of a high altitude (being more particularly abundant from 
about 2000 to 5000 feet above the sea) , [had always thought 
it extremely improbable that it could be absolutely con- 
specific with the particular form (so much resembling it) 
whose manifestly normal range is the vineyards and cul- 
tivated grounds of the lower districts ; yet the differences 
were so slight between the two, and both forms were so 
inconstant, that it was difficult to arrive at a satisfactory 
solution of the problem. But, taking their habits again 
into consideration, I am inclined to believe now that the 
