74 



older one (Cat. 53) has a length of 15 mm., of which 5 mm. is reckoned for the nectosome, 

 10 mm. for the siphosome. 



The table o-iven above shows first all the Catalogue-numbers in one column, then the 

 amount of siphons, as they develop themselves on the stem, the primary .siphon being the 

 oldest, the second siphon coming next and so on. The Roman numerals indicate the kind of 

 tentilla which accompany the siphon in the same column. We will describe the four types of 

 tentilla further-on. 



The first specimen Cat. i']6 H. could not be figured as it had undergone too much 

 alteration by preservation. It consists of a small pneumatophore which is colourless, a few buds 

 of nectophores, one primary siphon, situated nearly immediately below the pneumatophore, necto- 

 and siphosome being hardly differentiated. At the base of the siphon is a group of tentilla, 

 which are still very young. Some palpons and other buds are already developed. 



The next Cat. 16 ^ D. (PI. XI, fig. 91) is better preserved. (Length: top of the pneuma- 

 tophore to aperture of the siphon 37., mm.). The pneumatophore is damaged, the top is slightly 

 pigmented. One bud of a nectophore («) is already much developed, as we .see the gelatinous 

 substance developed on both sides. Other buds of nectophores [bn) are smaller and not yet 

 differentiated. The nectosome and also the siphosome are hardly distinguishable one from the 

 other; this shows how very young our Cat. 163 D. is. The greater development of the necto- 

 phores in such young stages shows a difference from the larva of Crystallodes described 

 by Haeckel 69. These possess already a few mature bracts before the first nectophore is 

 to be seen. 



The primary .siphon [prs), the very oldest one, lies immediately underneath the pneuma- 

 tophore. It is mature and bears at its base a group of very young tentilla. They are similar 

 to those of Cat. 25 A. which we sketched (PI. XII, fig. 92). These tentilla belong to type /, 

 they are the smallest; they are situated in a great number near the base of the stem. We never 

 saw any tentacle fully extended; it probably lies quite coiled together and hidden beneath the 

 pedicles of the tentilla. The tentillum itself consists of a certain number of big cylindrical, 

 elongated spindle-shaped nematocysts; one and a half or two turns of the cnidoband, the 

 whole surrounded by a closed involucrum and of the characteristic lateral filaments and median 

 lobe. All the different parts of this tentillum are very tiny and fragile, the lateral filaments are 

 never curled up. These tentilla of t\pe I are not identical with those drawn by Dan.\ 58 and 

 Haeckel 69 and 88b and other authors do not describe them. 



Cat. 163 I), shows furthermore four mature palpons with their palpacles. 



The following numbers Cat. ji B. (PI. XII, fig. 96) and 51 A. are somewhat larger. 

 The pneumatophore is very much developed, and shows the most beautiful pigmentation towards 

 its top, the purple stripes issuing on the lateral walls of the spherical pneumatophore. Fewkes 

 82 only describes a similar pigmentation in his Agalnia papillosum. But in his specimen the 

 relative size of pneumatophore to the other appendages is not so strikingly different as in ours. 

 Perhaps Agalma papillosiim also belongs to the genus Crystallomia. 



In Cat. 71 B and 51 A. the appendages are the same as in 163D. In 71 B. the necto- 

 phores and buds of nectophores, the primary siphon with its tentilla type I and palpons 



