During the purification of commercial dimethoate, a fraction was separated which potentiated the rat oral toxicity of pure dimethoate. Addition of 2-10 per cent of this fraction to pure dimethoate potentiated its toxicity (rat oral LD5Q) by 3-4 times, a finding which at least partially explains the difference in toxicity between the pure and commercial material. This fraction also potentiated the toxicity of malathion, and O,O-dimethyl S-carbamoylmethyl phosphorodithioate to the rat. The potentiator appears to be effective only with certain species, since the commercial material was more toxic than the pure compound to rats, guinea pigs, rabbits and hens,, but not to mice, pheasants, locusts, aphids and flies. With rats the potentiator was active by the oral and intra-peritoneal routes but not following dermal administration. On demethylation of di-methoate by reaction with salts of O,O-dimethyl phosphorodithioic acid, O,O,S-trimethyl phosphoro-dithioate and a potentiating material were formed. The potentiator, also formed by the reaction of dimethoate with lithium chloride, may be a salt of O-methyl S-(N-methylcarbamoylmethyl) phosphoro-thioic acid or a closely related compound. On analogy with the potentiation of pure dimethoate by O-ethyl O-j9-nitrophenyl phenylphosphonothionate, tri-o-cresyl phosphate and tri-o-cresyl phosphoro-thionate10, it would appear that the potentiating impurity in commercial dimethoate might act by reducing the efficiency of the dimethoate detoxification mechanism acting through hydrolysis of the carboxyamide grouping. Methyl and ethyl 'Cellosolve' (2-methoxy- and 2-ethoxyethanol) have excellent solvent characteristics for the formulation of organophosphorus insecticides, particularly where a high degree of systemic activity is desired after application to foliage. As a result of routine toxicity checks on the properties of technical dimethoate formulated in methyl 'Cellosolve', an unexpected toxic hazard was observed. The rat oral ZJD50 decreased on storage from an initial 150-250 mgm./kgm. to 30-40 mgm./kgm. after seven months storage in England. After nine months storage under tropical conditions the LD5Q was less than 15 mgm./ kgm., and after storage at 100 C. for 63 'hr. in the laboratory the LD5Q was 8 mgm./kgm. The mammalian toxicity increased several-fold before any change in insecticidal activity was evident. Further Development of this unstable formulation was immediately stopped.Chromatography by described procedures7-9 of the decomposed dimethoate in methyl 'Cellosolve' yielded fourteen phosphorus-containing products. The degradation involved hydrolysis at the amide and all the ester groupings, a marked loss of thiono sulphur, and replacement of both the O-methyl and N-methyl -amino groupings by 2-methoxyethanol. No evidence was obtained for the formation of pyrophosphates or thionopyrophosphates. The most toxic fraction was purified and identified as an O,O-dialkyl S-(N-methylcarbamoylmethyl). phosphorothiolate with probably one, but possibly both, methyl groups of dimethoate replaced by 2-methoxyethyl groupings. This compound was less toxic to flies than dimethoate, but had a rat oral LD5Q of about 1 mgm./kgm. and a mouse intra-peritoneal LD5Q of about 0-5 mgm./ kgm. The increase in toxicity to mammals of the decomposed dimethoate formulation was greater than that attributable solely to this single toxic product. Since no other single component was toxic enough to account for the discrepancy, it appears that a potentiation of toxicity also occurred among the materials in the dimethoate - methyl 'Cellosolve' formulation. The relationship of stability to toxicity of dimethoate formulated in a number of solvents was investigated. Little if any hydrolysis or change in mammalian toxicity occurred in non-hydroxylic solvents. Polyhydroxylic solvents resulted in stability during short storage without the formation of highly toxic degradation products. Of several monohydroxylic solvents, only methyl and ethyl 'Cellosolves' resulted in formulations highly toxic to mammals on short-term storage at 70 C. From this work a commercial formulation was developed which showed no significant change of toxicity after 162 days storage at 40 C.The potential hazard which could have arisen is not unique for dimethoate, since O,O-dimethyl S-(N,N-dimethylcarbamoylmethyl) phosphorodithioate, O,O-dimethyl S-(l,2-dicarbethoxyethyl) phosphorodithioate (malathion), O-methyl O-(2,4,5-trichloro-phenyl) phosphoramidothionate (Dow ET-15) and O,O-dimethyl O-(2,4,5-trichlorophenyl)phosphoro-thionate (Dow ET-51) were also found to show a several-fold increase in mammalian toxicity after reaction with methyl 'Cellosolve' at 70 C. for eight days. When 'Cellosolves' are to be used in formulating phosphorothionate insecticides, the possible ester and amide exchange reactions to yield products of increased mammalian toxicity should be carefully investigated. A more complete account of these investigations will appear elsewhere.Publication of this communication is approved by the Director of the Wisconsin Agricultural Experiment Station, and by the Directors of Fisons Pest Control, Ltd. The investigation was supported in part by a Haight Travel Fellowship from the Graduate School of the University of Wisconsin to one of us (J. E. C.), and by grants from the U.S. Public Health Service, National Institutes of Health, arid from the U.S. Atomic Energy Commission (Contract No. ^T(ll-l)-64, project No. 14). We thank Dr. E. F. Edson, Dr. G. S. Hartley, Mr. A. J. Lambie and Mr. P. Carter of Fisons Pest Control, Ltd., and Mrs. Lydia McBride of the University of Wisconsin, for their encouragement and assistance.