114 HarozD Hearty, 
the same as in other molluscs, the front portion moving, in the act 
of collecting food, from a position represented in Fig. 3 through an 
arc that brings the forward teeth close to the entrance of the 
stomach (Fig. 19). In the second place it has been shown that from 
the point of entrance into the pharynx to their extreme anterior 
end the two rows of teeth separate along the mid line and the 
epithelial layer to which they are attached penetrates deep within 
the anterior portions of the radula supports. As the rows of teeth 
diverge and their long bases shift to the outher rounded surface of 
the supports their tips become more and more inclined outward 
(Fig. 11). This position they hold in a state of rest. 
As I have shown (HEATH, 1899) in Cryptochiton stelleri the teeth 
are not passively drawn toward the stomach in the act of feeding 
but as they glide by the mouth opening the sides of the radula are 
involuted thus causing the great major lateral teeth of the opposite 
sides to interlock. Essentially the same process takes place in this 
species. As the tongue sweeps past the mouth opening the tips of 
the teeth of the opposite sides are brought almost in contact their 
appearance being at this time as shown in Fig. 14. This movement 
is largely due to the great tooth adductor (No. 1) that as we have 
seen attaches to the bottom and sides of the deep epithelial fold 
penetrating the space between the forward ends of the radula 
supports. By the contraction of this muscle the epithelium, that 
seems to be somewhat loosely attached to the tongue in this region 
is drawn still deeper within the supports which brings the teeth on 
the inner curved face (of the supports) with their tips directed 
inwardly. From sections it appears certain that the supports them- 
selves are not wholly passive during this operation but undergo 
changes in shape, probably not very extensive, that also aid in 
producing the effect in question. 3 
NIERSTRASZ (1903, fig. 10) figures a muscle in Chaetoderma 
challengeri that appears to resemble the tooth adductor (No. 1) in 
this species. However he assigns to it the function of radula re- 
tractor so that its attachments must be different from what they 
are in the present case. A glance at Fig. 3 will show that even 
should the septum bounding the head cavity posteriorly be rigid 
(which it certainly is not) the contraction of muscle No. 1 could not 
by any possible play of the imagination cause the radula to swing 
into the position shown in Fig. 19. 
Muscle No. 2 in its action opposes No. 1. pulling the teeth and 
