GENERAL CONSIDERATIONS. 81 



although we should not be able to conclude this from the flexure of 

 the alimentary canal (Fig. 41 B). We must here briefly return to 

 the position of the anal aperture in the Ecardines. In Lingula and 

 Discina the anus lies on the right side of the body, in Crania, as 

 Joubin and Blochmann have proved, in the median plane at the 

 posterior end of the body. Blochmann is inclined to regard this as 

 a primitive condition. Since we, however, with Caldwell, regard 

 the posterior end of the body in the Brachiopoda (in view of the 

 conditions prevailing in Phoronis) as in reality belonging to the 

 ventral side of the body, we must consider the position of the anus 

 in Crania also as secondarily modified. As a rule, in the Brachiopoda, 

 the intestine is bent in the direction opposite to that which it 

 assumes in the Bryozoa (cf. Fig. 41 B, with Fig. 28, p. 57). The 

 external apertures of the nephridia also appear to have undergone 

 shifting in the ventral direction. Such changes in the position of 

 individual organs may, however, easily be explained by the crowding 

 of the body between the two shell- valves. Although these displace- 

 ments are a certain obstacle in the way of comparing the Brachiopoda 

 with the Phylactolaemata and Phoronis, the general agreement 

 between these forms (which is specially marked in the body-cavity 

 and the nephridial system) is so great, that we can hardly doubt their 

 close affinity. On this subject we refer the reader to the statements 

 of Blochmann. 



Since Steenstrup and others first pointed out that the Brachiopoda 

 are in no way related to the bivalved Mollusca (Lamellibranchiata), 

 repeated attempts have been made to ally them more nearly to the 

 Annelida, by laying undue weight on the agreement that exists 

 between the two groups with respect to the condition of the body- 

 cavity, genital organs, and nephridia in the Brachiopoda, and to 

 regard them, when possible, as Annelida adapted to a sedentary 

 manner of life. Morse thus calls them "ancient cephalized 

 Annelids," distinguishing them in this way from the sedentary 

 Polychaetes (e.g., Serpula), which he names "modern cephalized 

 Annelids." We must therefore refer briefly to those characters 

 which have been claimed as evidences of segmentation. 



The division of the body of the larva into three or four regions 

 is purely external, and, as far as is yet known, does not affect the 

 coelomic sacs. Balfour pointed out that the order of appearance of 

 the segments in the Brachiopodan larva differs from that prevailing 

 in the Annelida. In the latter, the first segments to form are cut off 

 from the posterior end of the body, while in the Brachiopoda the 



G 



