No abstract

and 1.15, for increasing molecular weight fractions, respectively. Furthermore, the reported uncertainty in individual Mw and Mn values is such that, in all cases but one, Mw/Mn could be as small as unity, giving a monodisperse fraction, with no violation of the stated precision. The one exception is the high molecular weight fraction, Mn = 735,000, for which the polydispersity effect well exceeds our experimental results at either extreme of the Ew/Mn ratio. This could possibly be explained if there were a high molecular weight tail to the distribution curve which might become lost in the base line without contributing to the apparent plate height. Further study is being conducted on this and related questions.

Visual observation of the freezing and melting of compounds in cells used for the determination of purity has uncovered some heretofore unexpected behavior. This behavior has been correlated with certain difficulties experienced in the measurement of purity, particularly when the sample is very pure. Means for partially reducing these difficulties are proposed and procedures for increasing the accuracy of purity measurements are described.

No abstract

Definite to strong carcinogenic activity has been shown by some monosubstituted and disubstituted derivatives of .1"3 The only active trisubstituted-DAB tested has been the 2',4',6'-trifluoro derivative.4 It seemed of interest to synthesize and test for rat hepatocarcinogenic activity all of the trimethyl homologs of DAB with Me groups in the primed positions only. These are all new compounds and can be prepared by the diazotization of the proper trimethylanilines followed by coupling with PhNMe2. The new azo compounds are listed in Table 1. Experimental SectionAll melting points were detd on a Fisher-Johns apparatus and are uncorrected. The C, , N analyses were performed in this department on an F and M Model 185 analyzer by Mr. Daryl Sharp.Where analyses are indicated only by symbols of the elements analytical results obtained for those elements were within ±0.4% of the theoretical values.Trimethylanilines. Mesidine,5 bp 224-228°, was prepd from nitromesitylene6 by reduction with Sn-HCl. 5-Aminopseudocumene,1 mp 62-63°, was prepd from 5-nitropseudocumene8 in the same way. 4-Aminohemimellitene,8 mp 29-27°, was obtd from 5-nitrohemimellitene,8 6-aminopseudocumene9 was prepd from 6-nitropseudocumene,10 and 5-aminohemimellitene,11 mp 75-78°, was prepared from 5-nitrohemimellitene8 by reduction (Fe-AcOH). 3-Amino-pseudocumene9 was produced by Fe-AcOH reduction of 3-nitropseudocumene which in turn was produced by the hypophosphorus acid reduction of the diazonium salt from 3-nitro-6-aminopseudocumene.12 2',4',6'-Trimethyl-DAB. Mesidine (60 g) was dissolved in a mixt of 113 ml of coned HC1 and 376 ml of H,0 and diazotized at 0°u sing 30.6 g of NaN02 in 150 ml of , . One-half hr after the final addn, a soln of 54 g of C6H5NMe2, 552 ml of 95% EtOH, 264 ml of HjO, and 109 g of NaOAc was added, and the soln was stirred for 24 hr. Extn with PhH and evapn of PhH left a semisolid material which