10 



MEMOIRS OF THE XEW YORK BOTANICAL GARDEX 



Gloeocapsa acervata sp. nov. 



Plate 2, figure 22 



Families very variable as to the number of cells they contain 

 as well as to their shape and size, having up to 256, probably 

 even more cells, spherical to slightly elongated and lobular, piled 

 together in heaps, adhering by their very soft, gelatinous walls 

 and forming a continuous stratum; cells spherical, 0.9-1.3 |j 

 diam., bright aeruginous, regularly arranged in the young fam- 

 ilies, or colonies, soon arranging themselves in no definite order 

 in the older families; tegument hyaline or almost transparent in 

 the younger colonies, changing to honey-color, or later to dark 

 brown, homogeneous throughout the colony. 



Growing on limestone between Hatillo and Arecibo, no. 1390, 

 type, and no. 1393 ; near Hato Arriba, Arecibo, no. 1425. 



It is questionable as to whether the above neAvly described 

 organism should be considered as belonging to the genus Gloeo- 

 capsa or to the genus Annci/stis. It is a borderline species. In 

 its very youngest stages, it conforms to the Gloeocapsa method 

 of development. The colonies of two, four, and eight are more 

 or less angular, the cells are symmetrically arranged, and in 

 some cases a slight stratification of the tegument may be seen. 

 The colonies soon become spherical, the cells are promiscuously 

 arranged, and with no indication of stratification of the tegu- 

 ment. They then resemble typical members of the genus Ana- 

 cystis. Their teguments are sticky, and the growing colonies 

 become agglutinated into masses with lobular surfaces, in this 

 stage superficially resembling Microcijstis ichiliyoblabe Kuetz. 

 There seems to be no especially visible modification of the cells 

 to form resting spores at the end of the life cycle. The tegu- 

 ments finally seem to become completely confluent, and the cells, 

 probably after a period of rest, begin actively to divide and new 

 colonies are formed in position. This seems to be a typical 

 Gloeocapsa characteristic, and for this reason I have placed the 

 material under that genus. 



Gloeocapsa livida minor var. nov. 



Plate 2, figure 23 



Families composed of 2 (rarely more) cells, more or less ag- 

 glutinated into small groups, forming with other algae, a more 



