Of great importance is the total absence in all 68 specimens of the kidney-shaped tentilla, 

 which have been described in other Agalinids by Leuckart 54, Vogt 54, Keferstein and 

 Ehlers 61, Claus 63, Metschnikoff 74, Fewkes 81, 82, in Crystallodes by Haeckee 69. 

 These tentilla are sometimes provided with bristles. H.\eckee 69 describes how these kidney- 

 shaped tentilla gradually grow into the tentilla with two lateral filaments and a median lobe. 

 He even sketches a tentacle (tentacle of a larva of the 27''^ day) where one sees the slow 

 gradation from one type to the other. This development of tentilla has never been found by 

 other authors, either in Crystailomia or in any other Agalmid. They always find that the 

 kidney-shaped tentilla develop first and that next to these another tentacle with trifid tentilla 

 appears. Keferstein and Ehlers even describe how they found small Agalma Sarsii (3 mm.) 

 where the only siphon already showed the development of a tentacle with trifid tentilla. They 

 are also the very first authors who give e.xpression to the idea that the primary siphon may 

 have fallen off. 



The)- say : 



"Wir mochten nach diesen Beobachtungen uns der Vermuthung Leuckart's anschliessen, 

 "dass manche Siphonophoren in der Jugend andre Nesselknopfe produciren als spater, und dass 

 "wenn zweierlei Formen Nesselknopfe vorkommen, man bei den alteren Polypen die Jugend- 

 "formen trifft wahrend die jungeren schon die endgultigen tragen". 



Ciiux 88 first thought with Haeckel that the tentilla of the primary tentacle developed 

 directly into those with terminal filaments and median lobe. He finds however that these 

 are not the definite ones, as in more advanced specimens of Crystallodes the tentacle of 

 the oldest siphon was still provided with the well-developed tentacle and trifid tentilla. These 

 tentilla, however, do not reach half the size of the definite tentilla. We too have found again 

 this difference in size. 



We distinguish four types of tentilla, of which type I (PI. XIII, fig. 98) is the smallest, 

 somewhat larger is type II, (PI. XIII, fig. 99, 100), the third in size is type III, (PI. XIII, 

 fig. 101), the largest are the tentilla of type IV (PI. XIII, fig. 102). 



Group I of Crystailomia spec, is, as far as the tentilla are concerned, an absolutely 

 rounded grou]j. Group II comprises specimens which diverge mutually, it is true, but which in 

 the main points show conformity. 



A detailed description of the tentilla is given in each group. 



Crystailomia Dana 58. 



45. Crystalhvnia spec, group I. PI. XI, fig. 91; PI. XII, fig.92 — 96; PI. XIII, figg. 103, 104. 



Stat. 86. Anchorage off Dongala, Palos-bay, Celebes. Cat. 53. formald. 4°/^,. One specimen. 

 Stat. 136. Ternate-ancliorage. Cat. 24 A., 24 D., 24 F., 71 A., 71 B., 88 E., 88 F., 88 K., 88 L., 



88 M., 88 N., 88 P., 151 C, 151 D., 151 F., 151 H., 175 C, 175 D., 175 F., 176 H., 



215 A., 215 L. formald. 4°/^. 21 specimens. 

 Stat. 141. Lat. i°o'.4S., Long. I27°25'.3 E. Cat. 223. formald. 4°/^. One specimen. 

 Stat. 144. Anchorage North of Salomakiee-(Damar-)island. Cat. 163 A., 163 B., 163 C, 163 D. 



formald. 4°/^,. 4 specimens. 



