2 ON THE AGAMIC REPRODUCTION AND MORPHOLOGY OF APHIS 
admitted that they do so in Adriophthalmia. But by this assump- 
tion we arrive at a still closer approximation of the different classes 
in regard to their cephalic structure; for all would, on this supposi- 
tion, have either five or six cephalic somites,—the former number 
being invariably met with in the true air-breathers (though in many 
purely aquatic forms also), while the latter is found only in those 
which respire by means of gills. 
I repeat, I can see nothing in this generalization but a simple 
expression of the facts. But I would go a step further, and add to 
this the Aypothesis, that in the Aréicvlata the head is normally com- 
posed of six somites, which are all fully developed only in Podoph- 
thalmia, Stomapoda,and some Branchiopoda , while in other Crustacea, 
some one or more of the pre-oral somites is more or less abortive, 
and in Arachnida and Jvsecta, the appendages of the first somite are 
sessile, and those of the second or third undeveloped. Admitting 
this hypothesis, I find further, that of the six cephalic somites, the 
sterna of three (the mandibular and two mawillary) are always 
situated behind the mouth and on the ventral surface of the body. 
The position of the other three varies; but the most anterior or 
ophthalmic is always bent upwards in consequence of the cephalic 
flexure, and not unfrequently, as in Insects, constitutes the greater 
part, or the whole, of the dorsal region of the head. The next two, 
or antennulary and antennary sterna, may present every variation 
from approximative parallelism with the axis, in Sguz//a, to extreme 
reflexion, as in /zsecta and many Crustacea. 
7. Nothing can be more variable than the number of the somites 
whence appendages are developed in the various classes and orders of 
the Articulata; and in the Jlpriapoda the total number of somites 
even is susceptible of an extreme amount of variation. But in the 
other classes it appears to me that there is a typical number of 
somites, from whence but comparatively few forms depart either by 
way of excess or defect. Thus, if we leave out the Leimodipoda, all 
Podophthalmous and Edriophthalmous Crustacea have twenty somites, 
of which six are cephalic, eight thoracic, and six abdominal. In a 
very few Lranchiopoda, and in Trilobita, there is more than the 
typical number of somites; but I believe that in all other Crustacea, 
where the number of somites is not twenty, it is less. 
The question of the typical number of somites in the body of the 
fusecta is one which has been much discussed. But all the theories 
on this subject with which Iam acquainted are, in my apprehension, 
vitiated by the mistaken view which their authors take of the com- 
position of the Insect’s head. Many seem to consider it to be a 
