PORIFERA. I. 



73 



to make this fact possible, a growth would have to take place involving- a complete transformation of 

 the form of the spicule, because the small sigmata are not inscribable in the flagellate ones; now it has 

 later been proved by Levin sen (Vidensk. Meddel. fra Nat. Foren. i Kobenhavn for 1893, i) that the 

 growth of the chelse exclusively takes place by apposition; of course the same fact must be supposed 

 to be the case with regard to the sigmata, so that the small sigmata and the flagellate ones are different 

 bodies without an\- connection with each other. Neither is any transition found between these two 

 forms, as also flagellate sigmata are found in the following species without being accompanied by 

 common sigmata'). I must add, however, that I have found no developmental forms of the flagellate 

 sigmata. 



Vosmaer (The sponges of the Villem Barents > Exp. 1880 — 81, Bijdragen tot de Dierkunde, 

 12. Af levering, 3. Ged. 29, PI. IV, figs. 35, 37, 38) mentions as Grllins vagabundus var. y a species with 

 both flagellate .sigmata and sigmata of the common form. To judge from the figures of as well oxea 

 as sigmata, and as the species has both kinds of sigmata, it is rather probable that the .species is 

 G.flagellifcr, but as Vosmaer gives no measures it cannot be decided with certainty. 



Locality: Station 52, 63" 57' Lat. N., 13" 32' Long. W., depth 420 fathoms. Only one specimen. 



Gcogr. distr. Off Marion Island in the South Sea, depth 50 — 75 fathoms (Challenger). The form 

 mentioned by Vosmaer is from the Barents Sea. Accordingly the species appears to be very wi- 

 dely spread. 



8. G. porosus Frstdt. 

 PI. XIV, Fig. 2 a — c. 



1887. Dcsiuncclla porosa Fristedt, Vega Exp. vetensk. lakttag. IV, 440, PI. 24, figs. 36— 37, PI. 28, fig. 15. 

 1896. Gcllius flagcllifcr Lambe, Sponges from the Atl. Coast of Canada, Transact, of the Roy. Soc. of 



Canada, Ser. 2, II, Sect. IV, 185, PI. I, fig. 4—4 d. 

 1896. 1 Gcllius flagcllifcr Topsent, Campagne de Caudan dans le Golfe de Gascogne, Annales de 



rUniv. de Lyon, XXVI, 280, PI. VIII, fig. 4. 



Tuberous, or rounded, more or less irregular. The surface shaggy; tlie dermal membrane thin. 

 The skeleton a somcivhaf irregular network zoith triangular or quadratic meshes. Spicula: Mcgasclera 

 oxea 0-3J — 0-J2"""; microsclera (common .sigmata?), flagellate sigmata o-oy — o-i""". 



Of this species we have one specimen, of an irregular, roundish form ; it is, however, somewhat 

 damaged, and perhaps it is only a fragment; by far the greatest part of the .skin is wanting. The 

 greatest extent of the specimen is ca. 30""". The colour (in spirit) is gray; with regard to consistency 

 the specimen is very brittle; it is perforated by numerous larger and smaller canals. The sponge is 

 throughout highl)- interwoven with sand and extraneous silicious bodies. The surface, as far as I 

 have been able to decide, seems to be slightly shaggy from projecting spicules. 



The skeleton consists of a rather irregular, unispicular network of triangular and quadratic 



I) In sigmata of the common form the fact is also that the smaller ones cannot be, while the finer ones may be, 

 developmental phases, as most of the smaller have their final form, and an apposition would only make them thicker; if an 

 apposition worth mentioning took place in the longitudinal direction, deviating forms would appear. For further particulars 

 of the growth of sigmata see under G. microioxa, p. 68. 



The Ingolf-Expedition. VI. i. ■'^ 



