286 FRESH-WATER ALG^E OF THE UNITED STATES. 



the diagnosis to suit their own observations ; they fail to agree. 

 Thuret hoped to have reached a definite limit, but neither Borzi, 

 Wittrock nor Nordstedt are in accord with him. 



Wittrock and Nordstedt suggest that the genus Anabcena be 

 divided into four subgenera under the following designations : 



Sub-Genus 1. TRICHORMUS, spores globose or subglobose, heter- 

 ocysts intercalated, distant from the spores. 



Sub- Genus 2. DOLICHOSPERMUM, spores subellipsoid, or sub- 

 cylindrical, heterocysts as in Trichormus. 



Sub- Genus 3. SPHAEROZYGA, spores subellipsoid or subcylin- 

 drical, heterocysts intercalated, proximate to the spores. 



Sub- Genus 4. CYLINDROSPERMUM, spores subellipsoid or cylin- 

 drical, heterocysts for the most part terminal and proxi- 

 mate to the spores. 



Entertaining doubts whether these forms are true, fully de- 

 veloped plants, we continue, provisionally, the old arrangement 

 as nearly as possible. 



That Anabcena-forms develop Lyngbya and that Cylindrospermum 

 and Mastigonema forms do likewise develop Phycochromes, has 

 been clearly elucidated in our observations of the living plant- 

 forms. A very satisfactory example was found on ponds of 

 stagnant waters, submarine, in the vicinity of Dennisville, New 

 Jersey. In all directions there were seen, July, 1880, large 

 floating masses, dull, pale greenish, tea color, or olivaceous, 

 gelatinous scums, composed apparently of two distinct filamen- 

 tous plants, the one moniliforin, the other with even margins. 

 The former is an Anabcena although not strongly marked by 

 heterocysts or spores, nor strictly in accord with any described 

 forms, nevertheless appears to be a form of the genus. 



Plate CXCV1II, figs. 9, 10, two filaments, the one in normal, 

 moniliform condition, the other with the cells in the process of 

 dividing : figs. 11, 12, 13, progressive stages of development ; 

 cells becoming square, adjoining walls resorbing and making a 

 continuous, homogeneous filament ; figs. 14-19, cytioplasm di- 

 vides, and forms septa ; articulations at first equal then dividing 

 into two, four, or six to a diameter. For details of this plant 

 see Lyngbya vermicularis. For reference to other developments 

 see Gylindrospermum. 



ANABJENA FLOS-AQUAE, Kg. 



Free swimming, submembranaceous, aeruginous; trichomes 

 more or less curved, often circinate. joints spherical, or 



