Journal of Applied Microscopy. 



•239 



(^^ 



■? 



of the presence of less foreig-n matter. The heat should be just sufficient to 

 boil the liquid, othenvise it is liable to froth and run over. The liquid is boiled 

 for one-half to three-quarters of an hour, or until the agar has completely dis- 

 solved. The jar and contents are now weighed, and distilled water is added to 

 replace that lost by evaporation. 



The jar is then placed in a water-bath which 

 is kept at a temperature of 50 to 60 degrees C. 

 for about two hours. The finely di\-ided in- 

 soluble matter gathers into masses and settles to 

 the bottom, leaving a clear, transparent liquid. 



The filtration of a two per cent, agar through 

 paper is usually a very slow, tedious, and vm- 

 satisfactor}- process. The author prefers filtra- 

 tion through absorbent cotton, inasmuch as this 

 can be done rapidly and well. A liter of agar 

 may be filtered in a few minutes, and frequently 

 it is obtained p>erfectly clear \\-ithout any deposit. 

 At other times a ver\- slight deposit may form in 

 the tubes, which, however, interferes in no w-ise 

 with the usefulness of the medium. The method 

 employed is as follows : 



The filtering arrangement is shown in Fig. 3. 



The receiving flask (l^^ liter capacit}-) is 

 connected with a Chapman aspirator. A large 



funnel provided with a rubber stopper is fitted to the flask. A Witte's per- 

 forated porcelain plate (10 cm. in diameter) is placed in the fvmnel and 

 steadied in position by the glass rod which passes through the central open- 

 ing. A disc of muslin having the same diameter as the plate ^^or a trifle less) is 

 placed in the latter. Then a circle of absorbent cotton, about 11 cm. in diame- 

 ter, is placed on top. and finally another layer of musUn. A second Witte's plate 

 is now placed on top and remains there as a weight, only during the preliminar\- 

 wanning up of the funnel. About a half liter of boiling water is then poured 

 into the funnel. The pump is set into action and the hot water rapidlv passes 

 through. It is returned to the funnel several times in succession, thus warming 

 up the fimnel, filter and flask. During the operation care should be taken to 

 see that the cotton closes tightly over the edge of the porcelain plate and that 

 no openings exists. While the pump is still active the upper plate is removed. 



While the filter is still hot and the pump active, the well sedimented agar is 

 slowly poured into the center of the filter. The liquid passes through as rap- 

 idly as it strikes the surface. The impurities are brought last upon the filter. 

 If the filtrate is not perfectly clear, when examined in a test-tube, it should be 

 passed through the same filter a second time. It is imponant that the agar 

 be well sedimented previous to filtration. If this is not the case, the liquid will 

 not pas> through perfectly clear, and the filter, moreover, will tend to clog. If 

 clogging occurs, the upper muslin layer may be carefully removed, in which case 

 the filtration will recommence. 



