No. 3.] THE EMBRYOLOGY OF THE EARTHWORM. 403 
The embryo rapidly elongates, becoming successively ovoid, elon- 
gate, and finally vermiform. Upon the full establishment of 
the germ-bands the teloblasts and cell-rows have a definite 
and characteristic arrangement which is retained throughout 
the development. This arrangement (Figs. 47, 48, 49, 63) is 
essentially the same in the three species of Luméricus I have 
examined, and an exactly similar arrangement exists in the 
embryos of a small fresh-water Oligochzte which I have unfor- 
tunately not been able to determine, but which is almost cer- 
tainly not Lwmbricus. The primary mesoblasts lie behind and 
somewhat above a horizontal plane passing through the middle 
of the body (Fig. 38), in contact with each other in the median 
line. New cells are always formed at the postero-lateral angles 
of these teloblasts (Fig. 60), and the mesoblastic rows curve 
thence forwards and downwards around the teloblasts, nearly 
or quite meet in the middle ventral line, and then pass outwards 
and forwards, soon widening into a pair of flattened plates in 
which the coelomic cavities are developed long before their 
concrescence on the ventral side (Figs. 63, 69). The neuroblasts 
are much further forwards than the mesoblasts, and le just 
opposite the lower (mediad) edge of the mesoblastic plate. The 
two nephroblasts lie side by side somewhat further back and 
a little towards the dorsal side. The lateral teloblast, which 
I have found in early stages of L. communis only, lies just in 
front of the nephroblasts opposite the upper edge of the mes- 
oblastic plate (Figs. 61, 67). 
The growth of the germ-bands keeps pace with that of the 
embryo generally. The teloblasts are carried steadily backward, 
each leaving behind it a trail of cells derived partly by continual 
divisions of the teloblast, partly by subsequent divisions of the 
cells thus formed. After the most careful study of a large 
number of germ-bands in surface view, and in transverse and 
longitudinal sections, I can state positively that the several cell- 
rows — mesoblastic, neural, nephric, and lateral —are wholly 
derived from the corresponding teloblasts, and in no other way. 
I have repeatedly observed all stages of division in all of the 
teloblasts, in the cells derived from them, and in the outer ecto- 
blast. The division-figures prove beyond question that the cell- 
rows elongate in the manner described, and I have never seen 
the least indication of any other mode of increase. 
