CALCAREOUS ALG^ FROM MURRAY AND COCOS-KEELING ISLANDS. 293 



They are usually empty, so far as observed, but in the few that show proto- 

 plasts, no regular divisions have been noted. 



The conceptacles are very numerous in the original materials from New 

 Guinea described by Heydrich, and are moderately abundant in certain 

 parts of the specimens from Murray Island. In the many microtome 

 sections that the writer has made from decalcified material, both of the New 

 Guinea and the Murray Island specimens, the conceptacles have been found 

 either empty or with disorganized or fragmentary inclusions that have given 

 no certain clue as to the nature of the conceptacles. Heydrich assumes that 

 they are cystocarpic, but fails to give any description of their contents. 

 None of these cavities shows any traces of the tuft of central paraphyses 

 that have been supposed to be characteristic of at least the tetrasporangial 

 conceptacles of the genus Lithophyllum, and, on the other hand, none of them 

 shows the elevated cylindric caducous ostiole that characterizes many of the 

 species currently' referred to Goniolithon. But the West Indian Goniolithon 

 boergesenii Foslie has no such elevated ostioles — in fact, it has conceptacles of 

 very much the same external appearance as those of G. orthoblastum, though 

 smaller. The zonated appearance of a cross-section of the thallus, due chiefly 

 to interpolated secondary hypothallia, regularly or irregularly developed, is 

 more characteristic of the genus Lithothamnium than of the genus Goniolithon, 

 but the general arrangement of the cells, their firm walls, and the presence of 

 scattered " heterocysts" in the perithallic layers induce the writer to place 

 this curious plant in Goniolithon rather than in any other of the now recog- 

 nized genera. 



In its highly developed and large-celled hypothallium the later-described 

 Goniolithon megalocystum Foslie,' from the Karkaralong group of islands of the 

 Dutch East Indies, is somewhat suggestive of G. orthoblastum. The writer 

 knows this from description, photographs, figures, and a minute fragment 

 of the type; its hypothallic cells would appear to be smaller (25 to 40 ix by 10 

 to 18 jit) J and its conceptacles (sporangial) much larger (1.5 to 2.0 mm.); it 

 has no frequently alternating zones of perithallia and secondary hypothallia, 

 and it shows no layers of hypothallic cells with intracellular papillae from 

 their basal walls, which constitute such a striking and peculiar feature of G. 

 orthoblastum. 



The genus Goniolithon, as first- proposed by Foslie, differed fromthe Goni- 

 olithon as emended by him two years later' in the diagnostic characterization, 

 in the specified type, and in general species content. The designated type of 

 1898, Goniolithon papillosum (Zanardini) Foslie, was in 1900 placed by Foslie 

 in Lithophylhan and the name Goniolithon was then transferred to another 

 list of species with type unspecified but with G. bras sic a-florida (Harvey) 

 Foslie standing first. If this transfer of G. papillosum to Lithophyllum 



'Siboga-Exped. Monog. 6i, p. 48, figure 20 + plate 9; figures 8, 9, 1904. 



^K. norske Vidensk. Selsk. Skr., 1898, No. 2, p. 5, 1898; 1898, No. 3, p. 8, 1898. 



'K. norske Vidensk. Selsk. Skr., 1900, No. 5, p. 15, 1900. 



