SY R. J. TiLLTARt). 91 



The c"?6^ic/6' averages about 9/x in thickness, but the tough outer 

 stratum is usually not more than 2/x in thickness. Along the 

 mid-rib, this outer stratum becomes thickened to 4/x or 5)u,; it is 

 also somewhat thickened along the dorsal and ventral edges of 

 the gill. Spines are present upon the mid-ribs and edges of the 

 gill, but usually only from the base to some definite point, mark- 

 ing the end of the original basal joint of the gill. 



The hypodermis is not modified into columnar or spindle- 

 shaped cells as in the LestidcB, but remains everywhere a flat 

 epithelium, with the nuclei lying parallel to the cuticle. There 

 is, however, a very distinct dilierence in size between the nuclei 

 found in the basal joint of the gill, and most of those found in 

 its more distal portion. The former average about 8/x in length, 

 and are even larger (up to 10/x) in those cells from which the 

 principal internal laminae are developed; the latter only average 

 6/x in length, and do not stain with the same intensity. In 

 many places, the hypoderm-layer is exceedingly thin, only 3-4 /x 

 in thickness, so that the nuclei, even though lying parallel with 

 the cuticle, cause periodical bulgings in the hypodermis. In 

 spite of the narrowness of the gill-blade, many of the cells of 

 the hypodermis seem to take part in forming alveolar tissue; if 

 the gill be stained in toto, this meshwork can be seen lying 

 everywhere within the basal portion of the gill (except at the 

 extreme edges), and even in the more distal part, the meshwork 

 can be seen running along both sides of the rachis, nearly to the 

 tip. 



In some gills, hypodermal pigment is almost absent; in others, 

 the pigment may occur either in transverse bands, or in irregular 

 patches. A few are deeply pigmented all over. 



The tracheal system varies much in details, but the branches 

 never come oli" from the main stems perpendicularly (as in 

 Lestince), but always obliquely. 



In naming and describing the sub-groups into which I propose 

 to divide these gills, I have had in mind their descent from an 

 original two-jointed gill. Four phylogenetic stages can be seen 

 in this process, and I have classified the gills accordingly, as 

 follows : — 



