270 



different developmental stages are present. The numerous round 

 holes in the plates are all smooth, and there is no articulation 

 developed in the posterior end of the plates. The shape of these 

 plates when fully developed seems to be quite as that of the smaller 

 plates found more posteriorly in the specimen (Fig. 51. 5). These 

 plates are fully developed, and they have a distinct margin. The 

 holes are irregularly dented and in the middle as well as in the 

 articular end of the plates a thick network is developed. The length 

 of these plates is ca. 750 // by 600 jti in width. In the posterior- 

 most end of the specimen there are found some relatively small oval 

 plates, measuring only 550 fi in length and 375 ^ in width. Their 

 outline is not smooth and the holes are dented. In the middle 

 and in the anterior end there is a network. 



Fully developed anchors are rare, and only found corresponding 

 with the large anchor-plates in the posterior end. They measure 

 ca. 1100 ft in length and 700 p in width. The anchors are usu- 

 ally faintly asymmetrical. Their arms are distinctly serrate, the 

 vertex is smooth and the stock is finely dented. The anchors cor- 

 responding with the large plates in the anterior end of the frag- 

 ment are, as the plates, not fully developed, and for that reason 

 their arms as well as their stock is smooth. 



All over the body there are two different sorts of rods. The 

 smaller are found in the skin, where they lie in a dense layer. 

 They are regularly oval (Fig. 50. 10) and measure only 8 — 10 fi. 

 The other sort is found in large quantities in the longitudinal musc- 

 les, but also in the circular muscles they may be found. They are 

 as the smaller ones regularly oval (Fig. 50. 6) but measure 15 — 20 fn 

 in length. 



Protankyra javaensis is characterized by the shape of the ciliated 

 funnels and by the size and shape of the calcareous bodies. It is 

 not closely related to any of the at present known species of Prot- 

 ankyra, except rodea Sluiter. This latter species is on the other 

 hand so insufficiently described, that the real relation between it 

 and javaensis cannot be ascertained. The two species are seemingly 

 easily separated by the shape of the anchors and plates, but as 

 SI u iter's figures are rather poor, the two species may ultimately 

 prove to be identical. For the present I think it the most reasonable 

 to regard them as different species. 



