378 INVERTEBRATE MORPHOLOGY. 



as appendages, thoiigli the absence of a corresponding nerve- 

 ganglion tells very strongly against such an idea. The ante- 

 rior portion of the digestive tract arises in the embryo as an 

 ectodermal invagination, and is frequently lined throughout 

 by a chitindus cuticle. In the higher forms (Malacostraca) 

 the posterior portion of this foregut is enlarged to form a so- 

 called stomach (Fig. 168, s), in which the chitinous lining 

 thickens to form a complicated arrangement of teeth, which, 

 moved by special muscles extending from the stomach to the 

 walls of the body, serve for the comminution of the food. 

 No salivary glands occur. The midgut is frequently of very 

 small extent, and has usually connected with it a digestive 

 gland (Fig. 168, I) consisting either of from one to four pairs 

 of simple or but slightly branched coecal tubes, or else of a 

 much-branched compact gland opening into the intestine by 

 two or more ducts. The hindgut {i), like the foregut, arises 

 as an ectodermal invagination, and is usually lined with chitin 

 and unprovided with special glands. 



The nervous system presents a typically metameric condi- 

 tion throughout the greater portion of the body, a pair of 

 ganglia occurring in each segment, united by paired connec- 

 tives with the ganglia of the preceding and succeeding meta- 

 meres (Fig. 168, vn). In the anterior portion of the body, 

 however, as well as posteriorly, a certain amount of concen- 

 tration and fusion of the various ganglia occurs. An ideal 

 condition in which no fusion has taken place would show a 

 pair of cerebral ganglia (Fig. 169, ce) with which more or less 

 complicated optic ganglia are connected. From the cerebral 

 ganglia connectives pass backward and unite with a pair of 

 ganglia {g^), clearly indicated in the embryos of many of the 

 higher forms, but not yet definitely known in the Entomo- 

 st'raca, though it seems probable that they occur in these 

 also. The metamere and appendages which should properly 

 be associated with them seem to have disappeared ; that is 

 to say, they are the sole representatives of a metamere inter- 

 vening between the cerebral and antennulary segments. These 

 ganglia are united by a pair of connectives with a third pair 

 sending nerves to the antennules (gr"), and these again with a 

 fourth pair belonging to the antennary metamere (g^), and so 



