445 
persistence dorsally is not required and the muscle atrophies there in 
order that the body wall may not be inconveniently thick. 
A fact connected with the muscular layers, which has hithert o 
been unnoticed, is that the longitudinal layer is sometimes external to 
the circular layer. When this obtains (I have noticed it in about 6 
of my specimens), both layers vary from the normal type. The longi- 
tudinal layer is no longer, or only slightly, bi-pinnate (see fig. 2), and 
is divided into two parts. The outer retains some similarity to the 
usual form, but has much stouter fibres; whilst the inner has each 
fibre surrounded by a muscular sheath, so that the appearance is pre- 
sented of large cells with prominent irregular nuclei. The outer pre- 
dominates dorsally, and the inner ventrally, but sometimes there is 
only one layer dorsally and one ventrally. The circular layer is always 
enormously developed. It is difficult to see how these occurrences can 
be regarded as individual variations, and although all were carefully 
identified as L. herculeus, I suspect that the differences are specific, 
and that, when the various species of earthworms are all carefully 
examined with the microscope as well as with the naked eye, some 
more species will have to be created. 
The following measurements illustrate the relative thicknesses of 
the respective layers, the ratio of one layer towards another being 
fairly constant: 
ZeiB, oc. Microm. 3, obj. ay. 
clitellum, including cuticle -65 mm 
longitudinal muscular layer -075 » 
Dorsal to chord { 
:725 » 
f 1 a clitellum ++» see er amelie ay 
Ventral to chor longitudinal muscular layer -4 » 
Da 
clitellum --- ur 
dr 5 
Dorsal to chord { longitudinal muscular layer -1 » 
+6 » 
2 aloni ee 
Ventral to chord longitudinal muscular layer -3 » 
si 
The circular layer, though more variable, also has its greatest 
development ventrally: 
Dorsal +- - +05 mm 
Li { Ventral --- rss 0 
f Ventral --- rid À 
2 
‘ \ Dorsal eee … +05 » 
