80 LEONELL C. STRONG 



this method it is impossible to differentiate between the be- 

 havior of a tumor that developed fast for the first few weeks 

 and then became practically stationary and one that showed a 

 slow initial impulse and a. rapid progressive advance during the 

 latter part of the experiment. A modification of this method 

 has been used by Tyzzer and others. By weighing the tumor 

 mass at definite intervals of time, it is possible to compute the 

 rate of growth, provided that the rate of size increase was pro- 

 gressive. It is, however, a matter of common observation 

 that transplantable (or even spontaneous) tumors do not pro- 

 gress uniformly. This method does not give relatively accurate 

 growth rates. 



By palpation one can estimate the relative increase (or de- 

 crease) week by week. The method is an advance over the tw^o 

 older methods, but has a few disadvantages. Tumor masses 

 are usually irregular in outline. The time element therefore 

 involved in computing the volume of such a mass is consider- 

 able. The method, moreover, does not take into considera- 

 tion that there may be necrotic or hemorrhagic areas present 

 in the tumor mass. 



By weighing of Hype' masses from 0.01 gram to 10 grams we 

 have endeavored to eliminate the disadvantage of the previous 

 methods. The entire history of the mouse under consideration 

 was kept on one chart of coordinate paper. By comparing 

 each individual mass (determined by palpation) with the series 

 of type masses, we were able to estimate the approximate weight 

 of the tumor under investigation from week to week. Calcula- 

 tions of rates of growth were only made on tumors which 

 were firm. The method is not as accurate with recurrent masses 

 for the reason that ulceration and hemorrhagic areas are ap- 

 parently more prevalent than in the original inoculated growth. 

 The data thus obtained were used in the determination of rates 

 of growth (both progressive and regressive), studies on viru- 

 lence, etc., as outlined below. The method gives only an 

 approximation to the true state of affairs. By the use of large 

 numbers of mice, however, closer approach toward accuracy 

 can be made than by any other practicable method. 



