432 THE ALIMENTARY CANAL OF THE IMAGO. 
that the convoluted gland tubes are developed, and they remain 
connected with the cesophagus, as Weismann represented them 
[2, Pl. IX., Fig. 15], for a long time. 
Hence, I think it is probable that the new gland tubes are 
developed from cells formed by proliferation from the blind end 
of the duct, just as the gland tubules originate in the mammary, 
salivary and lachrymal glands of Vertebrates, as solid cell growths 
from the Malpighian layer of the epidermis. The invaginations 
which form the distal extremities of the ducts either appear 
at a subsequent stage or are comparatively short when the 
gland is first formed; the latter condition has apparently led to 
the received view with regard to the manner in which the 
tubular glands of insects are developed from the epiblast. 
Addendum.—When ihe above was already in type, my atten- 
tion was drawn to a very important paper by Paul Mayer* on 
the development of some decapod Crustacea. Mayer describes 
the development of the hind-gut, as Graber did from the blasto- 
poral invagination. In this he completely agrees with my views, 
except that he describes no proctodzeum in the sense in which 
I have used the term. Great confusion has arisen by the use 
of the words hypoblast and mid-gut with meanings which are 
special to works on insect embryology. If the mid-gut is ever 
developed from free yolk cells in Arthropods, hypoblast is not 
an appropriate designation for the layer of cells which forms 
its wall; nor can the term proctodeum be properly applied to 
a blastoporal involution. The application of the latter some- 
times to one and sometimes to another involution has led to 
erroneous generalisations. 
* Mayer, Paul, ‘Zur Entwickelungsgeschichte der Dekapoden,’ Jenaische 
Zeitsch. f. Naturwissenschaft, Bd. xi., 1877. 
