SIROSIPHON. 269 



ments so polymorphous in size and form, fear Uhis plant is 

 innocently responsible for the record of many species which 

 have absolutely no value. For illustration, looking at Plate 

 CXC, which represents S. pulvinatus, some of the diversi- 

 ties of form become evident. No diagnosis of this form, 

 nor of turfaceum (a synonym), by different authors, acknowl- 

 edges filaments of a single row of cells ; nevertheless it is 

 certain that all young forms are uniseriate, and that they 

 often grow to considerable length without dividing laterally. 

 Such condition represents S. torulosus, Rab., and #. saxicola, 

 Kg.; it is not unlike other young forms (figs. 5-8). >S'. minu- 

 tum, Hass., is a still more slender form which occurs also, 

 not infrequently mingled with the larger forms. 



8. crustaceus, (Ag.) Eab. ; 8. rugulosus, Kg.; 8. secundatus, Kg., 

 8. Heufleri, Menegh. ; 8. rMzoides. Breb. ; 8. hormioides, Kg. ; 

 and others, appear to be mere conditions, or stages of 

 development of the same species. 



Plate CXC, figs. 1-3, macrogonidia ( Gloeocapsa cells ) ; 

 figs. 4-8, earliest development of filaments from macro- 

 gonidia; figs. 9, 10, two elongated young plants with a 

 solitary series of cells ; figs. 11, 12, older forms, with 2-6-8 

 series of cells. 



SIKOSIPHON PULVINATUS, Breb. Forma ALPINUS, (Kg.) Wolle. 



Bearing some resemblance to 8. Alpinus, Kg., and with 

 habitat similar to the last ; it exhibits now and then a dis- 

 tinct, yellow heterocyst, and possesses here and there a 

 spine-like outgrowth, or attenuated apex, which is a young 

 growth with very short cells. 



Associated with this plant there are, usually intermingled 

 during the later Summer months, many solitary cells 

 (Gloeocapsa, etc.), which are spores or macrogonidia. They 

 may be found in many progressive stages of development. 

 They imply two processes of propagation, which may be 

 illustrated as follows: 



Plate CXCI, figs. 1, 2, are two ends of filaments with 

 young growths, spine-like processes ; fig. 3, a broken-off 

 part of a filament, the sporangia, or cells bearing the niicro- 

 gonidia, are enlarging and passing out of the sheath ; figs. 

 4, 5, the same increasing in size internally and externally 

 and producing such encysted cells as figs. 6, 7 ; the tegu- 

 ments of these dissolve gradually and then set the daughter 

 cells free ; these (figs. 8, 9, 10) pass through another similar 

 cycle of evolution, and again another and so on often for 



