rence 
pe 
1888.] 139 (Grote. 
family and subfamily terminations, and I am finally opposed to the bar- 
barous names used by Mr. Scudder for these groups in the butterflies. 
There is a certain amount of natural error which a student may fall 
into while gradually becoming acquainted with a large amount of new and 
differing species, as to which no work was before him, and through which 
he had to break a path. All things considered, no one in my position 
could have escaped having to change his views and cancel some of his 
work, Ihave always quickly acknowledged and corrected my mistakes, 
as all who have followed and used my previous writings, I think, admit. 
With these explanatory remarks, I would now offer a 7éswmé of my con- 
clusion and studies on the family. 
It must be acknowledged that the Noctwide are dificult of limitation as a 
family by exclusive characters. They may be shown to differ in turn in 
single points from other family groups of moths, but certain genera in 
every fauna are difficult to place. As to subfamilies, Lederer shows that 
these can only be defined comparatively, and not exactly, or, as he calls 
it, scientifically. The groups here recognized are merely tentative associa- 
tions of genera to which I have given a subfamily name ; they contain all 
of them genera which may be displaced by future enquiries, but they help 
the comprehension of the family and enable us to consider certain assem- 
blages together. As to their names, I have not followed any rule of 
priority ; Guenée gives some of them a family form. Ihave given them a 
uniform termination, and derived them from the most prominent genus 
they contain. 
The summer, that pulse of the year, the length of whose recurring beat 
ig at once the measure of the time elapsed since the culmination of the last 
ice period, gives us a prevailing northward direction for the winds that sweep 
the North American Continent. These offer serial paths along which num- 
bers of feathery-winged moths are hurried. We have wind visitors from 
the West Indies upon our shores during the whole season. Some of these 
become partial citizens by breeding here, others do not, and their lodg- 
ment upon our territory is precarious and accidental. The list of species 
known to visit us in this manner is already somewhat extensive, while the 
southern part of the peninsula of Florida is occupied permanently by the 
assemblage of tropical insects. This subject leads us to consider briefly 
the distribution of our Voctwida. 
The Geographical Distribution of the North American Noctwide must be 
studied in connection with the topography of the country and the range of 
the food-plants of the caterpillar. It is found that mountain chains afford 
the most eflective barrier to the distribution of species. Their presence 
explains the fact that Ohio insects are often absent in New York, or not 
go abundant on the north and east of the Alleghanies. <A study of the 
ranges and lateral branches of the Rocky mountains, as they are deline- 
ated, gives an idea of the different faunal provinces which are discovered 
to be more or less restricted to the valleys between the spurs. It is shown 
that, often at short distances in this region, the character of the moths in 
