AND EMBRYOLOGY OF LIMULUS. 15 



the teeth at the base of the limbs, is further triturated by the numerous hard teeth of 

 the crop, while the more nutritious fluid portions strain through the narrow passage of 

 the singular hollow cone. 



The inner walls of the stomach are destitute of chitin, the long, close-set, large papillae 

 being edged with a thick layer of columnar epithelium. The twelve circular folds of 

 pavement epithelium are also lined with a sunilar columnar epithelium. 



The four biliary ducts open into the intestine proper, which is lined as far as the 

 I'ectal folds with an epithelial membrane, is divided by longitudinal and transverse lines 

 into squares forming close-set, square, flattened papillae ; on the posterior half of the 

 intestine the longitudinal lines are more numerous than the transverse, the latter being 

 partially obsolete, so that the inner surface of the intestine is gathered into fine longi- 

 tudinal folds, the fii'ee edges of the folds being irregrdarly serrated. 



These folds consist of pavement epitheliiun (or mucous membrane), the free edges of 

 which are of columnar epithelium, the cells being long and narrow, while the nuclei are 

 not so large and distinct as in the proventricle. 



The intestine within suddenly contracts at the beginning of the rectmn, but becomes 

 - larger posteriorly to the vent ; the interior is thrown up suddenly into ten large folds of 

 unequal size, which become smaller posteriorly. These rectal folds have the same 

 muscular and epithelial layers as in the other parts of the digestive tract, but the cells of 

 the pavement epitheliimi, instead of being uniformly round, are in places iri-egularly 

 diamond or spindle-shaped, as in plate 5, fig. 6. The columnar epithelium of the rectal 

 folds is lined externally (in the rectal cavity or lumen) with a lining of a clear, structure- 

 less, somewhat chitinous membrane which stains purple with haematoxylin. It would thus 

 appear that the secreting surface of the stomach-walls is, owing to these folds and 

 large erect papillae, very much greater than in the intestine. We have seen that the 

 stomach, like the intestine, lacks the chitinous lining, and this,, together with the 

 histological identity of structure between what we regard as the stomach and the intestine, 

 may seem to some as opposed to the view that this region is the mid-gut, stomach or 

 arclienteron ; but the fact that it is divided from the intestine by a slight constriction, that 

 it lies in front of the biliary ducts, and that the appearance and gross anatomy of the lining 

 is unlike that of the intestine, coupled with the perfect continuity of structure in the 

 oesophagus and proventricle, are to our mind sufiicient arguments for the position we hold. 

 Moreover in the lobster the two capacious biliary ducts empty directly into the true 

 stomach or mid-gut, the small straight intestine beginning at some distance behind the 

 origin of these ducts. 



Thus w^hile the stomach and intestine of Limulus agree in the absence of the chitinous 

 layer, the rectum in its longitudinal folds and lining of chitine repeats in a degree the 

 structure of the oesophagus. 



Comparing the digestive canal of Limulus with that of the lobster or Decapodous 

 Crustacea in general, we find that the oesophagus and so-called stomach (what we call in 

 this paper crop or proventriculus), are continuous parts ; that the true mid-gut or stomach 

 has, like the intestine, no chitinous lining, though the rectum of the lobster, as we find on 

 examination, has long rectal folds (besides large square raised projections), and is through- 

 out lined with chitine. There is thus a general correspondence or homology between the 

 Decapodous and Merostomatous digestive or enteric canal. Unfortunately we have been 



