W. H. Coe — Anatomy of Cerehratulus lacteus. 489 



Rhabdocoela. The inner portion of each cell is prolonged into a 

 slender process, above which lies the nucleus as in the cells of the 

 body-epithelium. The fine granular protoplasm above the nucleus 

 contains minute particles of a transparent secretion. Between these 

 " rhabdite-cells ", and especially among their bases, are many less 

 elongated cells [a') which are more or less comjiletely filled with a 

 transparent secretion often collected into globules of various sizes. 

 The secretion of the proboscis-glands is exceedingly sticky, so that the 

 proboscis adheres to any object with which it comes in contact. This 

 organ may possibly aid the worm in securing its food. It is often 

 thrust out as if to explore the surrounding objects. 



In the anterior portion of the proboscis (Plate XII, fig. 11), the 

 epithelium is very simple and is made up of ciliated columnar cells 

 like those in the rhynchodseum (Plate XII, fig. 8). A transverse sec- 

 tion through this region of the proboscis shows only longitudinal 

 muscular fibres, among which the nerves [2'^n) appear as two crescent- 

 shaped masses similar in position to the lateral cords of the body.* 



In the middle portion of the proboscis (Plate XII, fig. 12), the 

 epithelium is much folded and when the proboscis is not extruded 

 the columnar cells partially overlap, so that in a section the epithe- 

 lium appears to be stratified (fig. 4). In this region the muscular 

 layers are very thick. Towards the posterior end (fig. 13) the 

 epithelial cells become still more slender than in the middle region 

 and among them are very few " rhabdite-cells." The cells are mostly 

 glandular and are filled with a transparent secretion collected into 

 little globules. 



The muscular layers are made up of muscular fibres which are con- 

 tinuous with the interlaced fibres of the head. These are gathered 

 up from all sides into compact and strong bundles as they pass, just 

 in front of the brain, into the proboscis. These muscles at first show 

 no division into layers but all run parallel with the long axis of the 

 proboscis (Plate XII, fig. 11). Gradually, however, a circular layer 

 arises just outside the nerves, which divides the longitudinal muscles 

 into an outer and an inner layer. As the circular layer (fig. 12, cm) 

 becomes stronger some strands of the muscle pass tangentially from 



*Hubreclit (1) describes and figures (Plate XV, fig. 3) a similar section from the 

 proboscis of Cerehratulus macroren^ although he states that the section was taken very 

 far back in the proboscis. He states also that the nervous plexus has given place to 

 two longitudinal nerve-stems and that the epithelium is flattened. It would be sur- 

 prising if the anterior end of the proboscis in one species should correspond to the 

 posterior end in another species of the same genus. 



