172 Journal of Agricultural Research voi. xvii. No. 4 



When properly prepared, the diluting solution should contain not 

 more than 10 particles to a X i"™- square field in cell No. 3, at a mag- 

 nification of 320 diameters. A definite portion of the solution to be 

 examined was introduced into this standard dilutant after the coarser 

 mineral matter had been removed. That was accomplished by dis- 

 solving I gm. of the original sample in 50 cc. benzol in a stoppered cen- 

 trifuge tube, allowing the solution to stand overnight (17 hours) and 

 centrifuging for i hour at a speed of 800 revolutions per minute. A 

 small portion of this solution was then drawn off from the tube at a 

 depth of 10 mm., and i cmm. transferred to a glass-stoppered graduate 

 and brought up to 10 cc. with the paraffin oil dilutant. By this means 

 a dilution of i to 5,000 of the colloids present in the original sample was 

 obtained. This was found to be sufficient in most cases, but in certain 

 bitumens where the dispersed mineral matter was in a state of extreme 

 subdivision a further dilution of i to 50,000 was necessary before the 

 colloid particles could be conveniently counted. 



METHOD OF COUNTING COLLOIDAL PARTICLES 



The samples of bitumens selected for examination were obtained from 

 the commoner types of road material, ranging from hard native asphalt 

 to lighter oils and containing varying amounts of colloidal matter. 

 After having been subjected to the preliminary treatment mentioned 

 above, one or two drops of the properly diluted solution were rapidly 

 transferred from the lo-cc. graduate to the cell by means of a i-cc. 

 pipette and covered immediately by a i8-mm. cover glass, using a 

 bluntly pointed wooden rod to expel all excess liquid and assure a 

 perfect contact between slide and cover glass. After the excess solution 

 had hardened sufficiently by evaporation and the slide beyond the cell 

 limit appeared perfectly clear and colorless, the cover glass was sealed 

 with a 30 per cent solution of boiled Canada balsam in ether applied 

 with a hair-line paintbrush (No. o). A photomicrograph of a part of 

 the mounted cell with cross-line micrometer scale magnified 320 diame- 

 ters is shown in Plate 19, B. 



In order to obtain consistent results, the cell and cover glass should 

 be microscopically clean before mounting and the dilutant examined 

 from time to time to allow for corrections in the final results.^ When 

 properly mounted the cell should be free from air bubbles and the 

 colloidal particles should appear under the microscope evenly dis- 

 tributed and in constant, though restricted, motion. In correct focus 

 these particles were clearly defined as brilliant points of light against 

 a dark background, but a change of focus resulted in the development 



' The cleaning was accomplished by first boiling slide and cover glass in concentrated sulphuric acid, 

 then rinsing in water, alcohol, and benzol, drying with soft cotton or silk cloth, and rubbing with optical 

 tissue paper until thoroughly clean. 



