LIMNOMEDUSAE 43 



with Russell's figure, though the gelatinous substance of the umbrella is not quite as thick in the 

 present specimens. The gastric peduncle and the nematocyst clusters on the mouth-rim are distinct. 

 In the smaller specimens the gonads are as figured by Russell, being somewhat laterally compressed, 

 wavy, and attached to the subumbrella along the greater part of the radial canals, from the base of 

 the peduncle almost to the ring-canal ; but in the larger specimens the gonads are considerably pro- 

 longed, terminating in a large, pendent sac, the length of which may be equal to the attached portion. 



The structure and arrangement of the marginal tentacles need some additional remarks. Apparently 

 there are two kinds of tentacles ; Russell designates them as large tentacles and rudimentary or dwarf 

 tentacles, the latter being regarded as 'true tentacles in an arrested state of development'. The margin 

 of the umbrella is divided into a number of swellings of nematocyst tissue, one at the base of each 

 tentacle, large and oval below the large tentacles, very small below the small tentacles. The relative 

 number of large and small tentacles will be discussed below, sometimes they are almost regularly 

 alternate. According to Russell the tentacles have ' a solid core of endoderm cells arranged in several 

 rows '. The term ' solid tentacles ', however, should be used only when the endoderm consists of one 

 row of cells ; when there are several rows it means that a central canal is present, though under certain 

 circumstances it may be so tightly compressed that the cavity is obliterated. As a matter of fact, in 

 this species the large tentacles are hollow, even with a distinct and sometimes rather spacious central 

 cavity, especially in the proximal portions. In the smallest ' dwarf ' tentacles the endoderm is uniserial 

 in the distal part, multiserial in the proximal part, and during growth the multiserial endoderm extends 

 farther outward towards the tip. Each large tentacle is provided with a terminal knob and several 

 transverse bands of nematocysts encircling the tentacle in its distal part; for most of its length, 

 however, the bands surround only part of the circumference, tapering in width towards the proximal 

 end of the tentacle, the innermost side being destitute of nematocysts. The nematocyst clusters 

 are on the abaxial side of the tentacle. In the largest tentacles the number of nematocyst clusters is 

 frequently about 35, in exceptional cases as many as 40. The basal part of the tentacle is adnata to 

 the exumbrella as described by Russell. 



The 'dwarf tentacles differ slightly from the descriptions by Bigelow and Russell. When quite 

 young they are slender and pointed without indication of a terminal knob, but later on a tiny knob 

 begins to appear, and as the length of the tentacle increases, a small cluster of nematocysts is deve- 

 loped abaxially a little inside the tip, then a second one, and so on. Already in the youngest stage 

 these small tentacles have some nematocysts scattered throughout their length (small ones about 

 30// in length), but as soon as the knob and the clusters make their first appearance, they contain 

 nematocysts of a much larger type, about 60 fi long. I have seen every transitional stage from tiny 

 ' dwarf ' tentacles with a pointed tip to fully developed tentacles with numerous bands of nematocysts. 



The small specimen, 5 mm. wide, examined by Bigelow had 16 large tentacles and 8 minute dwarf 

 tentacles, all of which were situated on the marginal swellings below the perradial and interradial 

 tentacles, whereas no dwarf tentacles were attached to the 8 adradial tentacles. In the slightly larger 

 specimens examined by Russell, the large tentacles were likewise 16 in number, each of the perradial 

 and interradial ones flanked by a dwarf tentacle, but there were also 16 small tentacles, one m the 

 middle of the space between every two large tentacles and each with a minute basal swelhng of its 

 own. This ideal arrangement may also be seen in larger specimens with a greater number of tentacles; 

 there is always one dwarf tentacle situated on each of the perradial and interradial swellings, whereas 

 all the other small tentacles are situated between the large ones. A regular alternation, however, is 

 rarely seen, because by-and-by small tentacles develop into large tentacles, and the larger the specimen 

 the more irregular is the succession and the relative number of large and small tentacles. When a 

 small tentacle has obtained a certain number of nematocyst clusters, a new small tentacle appears 



6-2 



