2o8 American Quarterly Microscopical Journal. 



the Protococcus, Pleurococcus, and Chlorococcum of botanists. The 

 larger cells are called sporangia, but they do not differ from 

 the others except that they have become more matured; these 

 eject the granules (microgonidia), which develop and repro- 

 duce. This process continues during the greater part of the 

 summer, so that by the autumn months the strata have become 

 quite extended, and the cells often crowd upon one another to 

 form thick crusts. If we collect from these strata in the cooler 

 months of October, or November, when there is more moisture 

 in the atmosphere, we find that the Protococcus and the growths 

 which were there in summer {a a a) are almost beyond recogni- 

 tion; they have conglomerated into irregular masses {bbb), the 

 individual cells are breaking into parts, some elongate and add 

 cell to cell, and here and there a long filament of Ulothrix {c c) 

 has developed. We examine another mass; in this the filaments 

 predominate. 



Illustrations like the foregoing might be multiplied. They 

 leave little doubt in my mind, that the many unicellular forms 

 recognized as plants, are of a very doubtful character. To say 

 that all of them should," at once, be wiped out from the cata- 

 logue of plants, might be too sweeping a demand. I merely 

 call attention to the facts as they came under my observation. 

 Bethlehem, Pa. 



ON TWO FORMS OF COxMPARATORS FOR MEASURES 

 OF LENGTH. 



BY PROF. W. A ROGERS. 



{Received March 2pth, iSjp.) 



The subdivision of a given unit into exactly equal parts, 

 is a problem of extreme difficulty, and the difficulty rapidly 

 increases with the length of the unit to be divided. The 

 measurement of the subdivisions, presents almost equal difficul- 

 ties, though of a somewhat different character. 



Two methods ofifer themselves for the solution of the prob- 

 lem: 



I. We may assume the smallest subdivision to be an aliquot 

 part of the entire unit, and then obtain that unit by successive 

 increments of this constant quantity. If the number of sub- 

 divisions is large, it will be found practically impossible to 



