SANGER. — VOLATILE COMPOUNDS OF ARSENIC. 137 



of the succeeding experiments. To try the action on paper, a tight 

 box was lined with sterilized paper colored by Schweini'urth green 

 and fastened by sterilized starch paste which had been inoculated 

 with mucedo. Air was drawn through the box for 39 days into 

 argentic nitrate which then gave a positive test for arsenic. 



The sensibility of the mucedo varies with different conditions : 

 humidity, amount of oxygen present (the formation of gas ceases if 

 air be lacking), and the quantity and quality of the arsenic compounds. 

 The best results were obtained in ground containing 0.01 to 0.05% of 

 substance, while 4 to 5% distinctly retarded the growth. Arsenic acid, 

 arseniates, or alkaline arsenites gave the best results. There can be 

 a tolerance for high doses established, however, if the mould is habit- 

 uated to a progressive increase. The action goes on better in solid 

 ground than in liquid, and the best nutritive material is a carbohy- 

 drate. In albuminoid matter there was but little action, whereas in a 

 mixture of albumen and glucose, it was intense. Whenever the 

 mucedo showed intolerance, other moulds grew which do not decom- 

 pose arsenical matter, e. g. P. glaucum. This explains the failure of 

 other observers (including myself), who used large amounts of arsenic 

 and yet obtained a quantity of mould. 



The action of other moulds was examined in the same way, but it 

 was necessary in certain cases to lessen the amount of arsenic. By 

 this means a mould corresponding to the properties of Aspergillum 

 virens was found to react. Very slight effects were obtained from 

 Sterigmatocystis ochracea, Cephalothecium roseum, and Mucor ramosus. 

 Finally, from a piece of carrot left in the open air, a new mould was 

 isolated which proved to be identical with the Penicillium brevicaule 

 discovered by Saccardo on decaying paper. Experiments on this 

 mould showed it to be capable of more intense action than any other. 

 With milk culture Gosio claims to recognize the odor from 0.02 mgr. 

 of sodic arsenite and thinks it may be still more delicate. In the 

 experiments with P. brevicaule, the absorbent was an acid solution of 

 potassic permanganate, but the strength as well as the subsequent 

 treatment is not given. The extreme sensibility of this mould led 

 Gosio to propose its use as a means of testing for arsenic in toxico- 

 logical work, and the second of his papers is devoted to the working 

 out of this method. 



The power of decomposing arsenical organic matter is proposed as 

 a means of distinguishing between two moulds of similar properties. 



As " arsenio-bacteria" Gosio specifies those which have a prompt, 

 intense, and lasting action. He thinks that all micro-organisms may 



