IN THE BRTTTSTI ISLANDS. 47 



to be marked by continuous bands. The eyes are naked ; antennas 

 generally simple ; body usually untufted ; abdomen smooth and taper- 

 ing; the dorsal surf ace rarely crested; the tibias not unfrequently armed. 

 Packard says that the epicranium is longer than in the Noctuince. The 

 basis of this sub-family is comparative form. No single structural 

 feature holds the genera together, and Lederer does not recognize its 

 existence. Nevertheless, the moths and larvaa seem to me more or less 

 readily distinguishable, and that we may retain the sub- family term. 

 Dr. Packard says (I.e.) that these two sub-families ' agree in the main 

 with the Trifidce and Quaclrifidce of Guenee, though the use which he 

 makes of the venation seems to us to lead to artificial distinctions. 

 The very constant venation of this family does not admit of any 

 variation in the grouping of these veins and their branches, and hence 

 they offer characters of secondary importance.' With this statement 

 my experience fully agrees. I do not think that the venation alone 

 should decide family position. There are already too many exceptions 

 known to the system of Herri ch-Schaffer, which is, in some cases, con- 

 sidered the test of family character by Lederer. Primarily this 

 sub-family falls into two groups : First, the tribe Catocalini, in which 

 the secondaries, though often gaily coloured, are covered by and sub- 

 ordinated to the primaries. The European genera belong chiefly to 

 this tribe. Secondly, the Pheocymini (Fheocyma, Homoptera, Erebus, 

 etc.) in which the secondaries are partially exposed and marked like 

 the fore wings, decidedly geornetriform moths. The larva thus first 

 becomes geometriform as we recede from the higher Noctuidce, and 

 then the perfect insect follows suit " (' Canadian Entomologist,' vol. 

 xxii., p. 109). He further writes : " Some idea of the preponderance 

 of Catocalince in North America may be given by the statement that in 

 Europe there are about fifty-six species belonging to about sixteen 

 genera, while in North America there are about two hundred and fifty 

 species belonging to about fifty-six genera. These latter figures may 

 be changed by new observations with us, but hardly diminished. In 

 the Deltoidince the proportions are more in conformity with the nume- 

 rical relations in the typical group, the Noctuince. The reason I have 

 given for this preponderance of the Catocalince, lies in the physical 

 geography of the Continent, the prevailing atmospheric and ocean 

 currents, all of which favour the introduction of southern or tropical 

 lepidopterous forms ; and we must consider the Catocalince as tropical 

 in general character in the same way as we consider the other groups 

 of the Noctuidce as belonging to temperate regions of the earth's sur- 

 face " (' Canadian Entomologist,' vol. xxii., p. 148). 



Of the Catocalim as a tribe of the Catocalince Grote further writes : 

 " In this tribe the secondaries are oftenest gaily coloured (mostly yellow) 

 and banded, still subordinated to the primaries which show, more or 

 less adequately, the usual Noctuidous ornamentation. The abdomen is 

 rarely tufted. On account of the shape of the primaries, the form of 

 the abdomen, the abdominal tuftings, the pattern of the wings beneath, 

 the approach to Ophideres, I regard the genus and species, EupartJtenos 

 mililis, 'Ann. N. Y. Lye. Nat. Hist.,' as entirely distinct from Catocala 

 sp. The most important genus is Catocala, in Europe with 22, in 

 North America with upwards of 100 species. A division of this genus 

 on the peculiarities of tibial armature (as I have suggested in Agrotis) 



