574 Notes in 1887 upon lepidopterous larve, ce. 
they are not hawk-moths. Their mode of flight, and 
especially their rudimentary and unused mouth-parts, 
are further points of affinity to the Saturnians. 
It therefore follows that the chief peculiarities of the 
Sphingide, as opposed to the main body of Bombyces,— 
the fact that they feed largely and are greatly specialised 
in relation to flowers,—are characters which were absent 
from their Bombyciform ancestors, and are still absent 
from Smerinthus, while they have been re-acquired com- 
paratively recently in the phyletic history of the majority 
of Sphingide. 
The most natural arrangement would be for the 
Sphingide to form the end of one special line of 
Bombyces, the order being the exact reverse of that 
given above. 
If, however, in view of their size and importance, it be 
preferred to begin the series of Heterocera with the 
Sphingide, the order within the group itself should be 
the reverse of that usually adopted in descriptive works, 
and should end with Smerinthus and the genera allied to 
it, and then the Saturnian genera should follow as in the 
list given above. 
I trust that these conclusions will stimulate entomolo- 
gists to carefully work through the life-histories of other 
species, so that the complete series may be constructed, 
and the affinities between the existing species of Bombyces 
may be expressed in the most satisfactory manner. 
4, THE CAUSE AND MEANING OF THE SPHINX-LIKE ATTI- 
tupE.—The extremely marked and characteristic Sphinx 
attitude assumed by the larve of Aglia, together with 
the fact that this attitude is most pronounced in the 
early stages of this genus and of Sphingide—when the 
larve habitually rest on the under sides of the leaves— 
led me to consider whether there might not be some 
causal relation between the position selected and the 
attitude assumed. Putting together all the facts, and 
collecting fresh observations, the following explanation 
suggested itself to me. In larve which assume this 
attitude the thoracic legs are not employed for the 
support of the body; hence, when the larve cling to a 
leaf or twig so that the dorsal surface or posterior ex- 
tremity is lowest, the weight of all parts anterior to the 
3rd abdominal segment is only indirectly supported by 
