(: *ebyair~ }) 
schuppen,” which are very numerous, are ovate, not always 
regular in size or shape, sometimes being evenly ovate, some- 
times tapering slightly towards the apex; the attachment 
stalk is long, tapering, slightly wider at its juncture with the 
scale, which is coarsely ribbed, several that I have examined 
having nine ribs, others however had fewer. The scales 
of the secondaries show no such modifications as do the 
primaries. From this unusual example I went for confirmation 
to Polyommatus dolus var. vittata, a species that has somewhat 
similar hair-like scales over the same area, only they are less 
copious and not quite so noticeable; in this case I found the 
long fine basal hair scales (bluish-white) more extensive, 
whilst the brownish ones (so prominent in menalcas) are 
perhaps less abundant but still very plentiful; they are 
similar in structure to those just described but rather finer; 
an extraordinary character however obtains in the ordinary 
blue wing scales, the whole of which are curled round so as 
to form more or less short tubes, the process appears to be 
that each side of the scales turns over, and occasionally they 
meet thus in the centre, but more generally one side will 
overlap the other and so form a more or less perfect tube, 
by this I mean of course that the basal and apical ends 
remain open—a tube that is sealed at each end naturally 
ceases to be a tube, becoming acylinder. This peculiar and 
interesting development obtains with nearly all the upper 
layer of blue scales with the exception of those in the terminal 
area, the lower layer in this case are not brown but are more 
or less transparent brownish-grey, and they (not being as 
numerous as is generally the case) retain their normal shape 
and position. The “ blasenschuppen” are plentiful, a long 
oval in shape, much the shape of a narrow specimen of the 
egg of the red-necked Grebe, with a longish attachment 
stalk as in menalcas, and having 7 or 8 rows of coarse reticula- 
tions, though in one large scale I counted 9, but the former 
seem to be the normal numbers. In the secondaries also the 
normal bluish scales are developed in exactly the same way 
into the more or less semitubular ones as in the primaries. 
Is it possible that we have here the early evolution of the long 
hair-like scales peculiar to this section of the genus Polyommatus ? 
