with buffer solutions is only partly affected by differences in structure.The low isoelectric point of fibroin (pH 2.2), although somewhat striking, is easily accounted for with a little knowledge of the structure. That the fibroin molecule contains an excess of free carboxyl groups over free amino groups has already been pointed out. Toa (5) demonstrated that fibroin was a negative colloid in aqueous solution and in an electrical field would migrate toward the anode. These observations are also confirmed by the fact that basic dyes form more stable combinations with silk than acid dyes.Literature Cited (1) Hardy,

the cooling which is preliminary to carbonation. By so doing, the yield is improved because otherwise oxygen from the atmosphere would be pulled in and react with the Grignard reagent. In a 0.2 mole run the yield of cyclohexanecarboxylic acid based on an 80% yield of cyclohexylmagnesium bromide was 83%; and the yield of acid based on a 92% yield of cyclohexylmagnesium chloride was 93%. It is recommended that the solution be diluted with ether (300-500 cc. in a 0.2 mole run) prior to carbonation, because of the sparingly soluble salt (RCOOMgBr) which is formed. Summary Cyclohexylmagnesium chloride and bromide can be prepared in excellent yields. In its reaction with magnesium, bromocyclohexane differs significantly from other secondary bromides. General recommendations are given for the use of RC1 in preference to RBr and RI compounds for the preparation of Grignard reagents.

mation or are at least considerably retarded after the CHsCl group has been "stabilized" by the presence of the two fluorine atoms on the adjacent carbon. The use of mercuric3 fluoride supplied the needed demonstration.This salt fluorinates C2H5CI easily but affects CH2CICHCIF very little, and CH2CICHF2 not at all. Similarly, it reacts quantitatively with C2H6Br, but not at all with CH2BrCHF2; it reacts violently with C2H5I but to affect CH2ICHF2, it must be heated for a long time at 160°, and yields only some 60% of CH2-FCHF2. It cannot be held that mercuric fluoride causes fluorination to proceed along different lines, as it reacts with CHBr8CHBr2 to yield successively CHBrsCHBrF -> CHBr2CHF2 (where the reaction stops when using SbF3), then CHBrFCHF2 -CHF2CHF2 while its reaction with CH2Br-CHBr2 yields successively CH2BrCHBrF and CH2BrCHF2, where the reaction stops just as it does with antimony fluoride.The fluorination of CH3CC13 gives successivelyThere is one other example of fiuorinating directly to a CF3 group, and that is C6H6CC13 -> C6H5-CF3. In both cases, the passage from the CC1F2 stage to the CF3 stage is rapid, an indication that the last chlorine had not been rendered inert by the two fluorine atoms. It should be noted that in both cases there is no halogen atom on the adjacent carbon.The "stabilization" hypothesis is in good agreement with experimental evidence derived from the interaction of fluorochlorides and zinc in (3) See following paper, This Journal, 88, 884 (1936). alcohol. From CC12FCC13, CC12FCCI2F, CHC12-CC12F, CH2C1CHC1F, CHC12CHC1F and CH2C1CC12F, it is extremely easy quantitatively to obtain CC1F=CC12, CC1F=CC1F, CHC1= CC1F, CH2=CHF, CHC1=CHF and CH2= CC1F, respectively. It is, however, quite difficult to obtain CC1F=CF2, CH^HF, CHC1= CHF and CH2=CF2, from CC12FCC1F2, CH2C1-CHF2, CHC12CHF2 and CH2C1CC1F2, respectively, and the yields are not good. It is almost impossible to obtain CF2=CF2 from CF2C1CF2C1, and even at 1000 pounds pressure, the yield is only 30-35%. Summing up the above observations the con-

EXTENSIVE study of gaseous detonation has been made by a number of investigators, but in spite of this fact the phenomenon is as yet not well understood. In presenting this paper, therefore, we will not attempt to explain all the phenomena with which it deals. The object of the paper is'to discuss the bearing of this factor on the operation of internalcombustion engines, and to describe the progress that has been made in controlling it by chemical means.Results of Early Study of Gaseous Detonation r In] 1881 Bertholet2 and LeChatelier3 independently discovered that the propagation of flame through mixtures of some combustible gases with air, and through mixtures of practically all combustible gases with oxygen in proper proportions, results in setting up a detonation wave.These results were confirmed by II. B. Dixon,4 who has carried out quite an extensive investigation of this subject. Mallard and LeChatelier5 noted that the development of the detonation wave is not progressive, but always instantaneous. They noted further that the detonation wave is characterized, not only by its great velocity of movement, but also by its intense luminosity. Dixon6 was also impressed by the sharpness with which luminosity is set up when detonation occurs.Bertholet and Vieille7 and also Dixon4 showed that the velocity of the detonation wave is constant, and Dixon advanced the theory that during detonation the flame travels at the same speed as sound at the temperature of the burning gases. Mallard and LeChatelier7 found that very large pressures are developed by detonation waves, but that such pressures exist only for an exceedingly brief period. This finding was confirmed by Dixon,4 who worked on the principle that, if a pressure is produced in a glass container greater than the glass will vrithstand, the vessel will be broken, although the pressure may endure only for a very short interval of time. Dixon gave a range of from 25 to 78 atmospheres for the magnitude of these pressures for various gases. He showed further that these instantaneous pressures are approximately four times the maximum "effective pressure" developed by the explosion.Gaseous Detonation an Important Factor in Internal CombustionExcept for some comparatively recent work, practically all the study of gaseous. detonation that has been made has been conducted in atmospheric tubes, or under other conditions that did not simulate those of internal combustion. From an economic standpoint, however, the detonation that occurs in internalcombustion engines is of great importance.8 Nearly everyone

A method has been described for the quantitative estimation of total tin in a variety of organotin compounds. The method consists of a preliminary decomposition by bromine in carbon tetrachloride followed by a nitric-sulfuric acid oxidation to stannic oxide.