Some Photochemical Experiments with Pure Chlorophyll. 349 



"be considered. From among the four substances at our disposal (methyl and 

 ethyl alcohols, acetone, and pyridine) we chose ethyl alcohol, for the following 

 reasons. Pyridine is the best solvent for chlorophyll and mixes very 

 readily with water, but on account of its strongly basic properties it would 

 be impossible to examine the influence of acids in its presence. As carbon 

 dioxide forms only a weak acid with water, the action of this gas on a 

 chlorophyll sol could not be examined in presence of pyridine. There is 

 also a difficulty in obtaining pyridine absolutely free from impurities, and 

 it gives oxidation products on exposure to light which interfere with tests 

 for formaldehyde. Also owing to its high boiling point it cannot be removed 

 by evaporation under reduced pressure. 



The use of methyl alcohol is also inadvisable, as tests showed that under 

 the conditions of our experiments it oxidises very readily to formaldehyde. 

 Similarly acetone was not available, as it also produces formaldehyde on 

 exposure to light. It must be emphasised that to the other sources of error 

 in the experiments of earlier workers dealing with the production of form- 

 aldehyde by chlorophyll this one, arising from the presence of traces of the 

 solvent used, must be added. 



Ethyl alcohol on exposure to ultra-violet light in quartz vessels gives 

 formaldehyde, but not when exposed in glass tubes. Therefore ethyl alcohol 

 was used in the preparation of the chlorophyll sol and glass vessels were 

 generally employed, quartz vessels being only used for special purposes. 

 When necessary the concentration of alcohol in the colloidal solution was 

 reduced by evaporation under reduced pressure.* 



Experiments. 



In these experiments a definite quantity of the chlorophyll sol was 

 introduced into test-tubes from which the air was subsequently evacuated. 

 The tubes were then filled with the particular gas or gas mixture desired, 

 sealed off, and exposed to light. When a mercury vapour lamp (3"5 amp.) 

 was used, they were placed at a distance of 10 cm. from the lamp. 



The results obtained with different gases were as follows : — 



* In this connection we should like to draw attention to a simple method, much 

 superior to the iodoform method in sensitiveness, for estimating small quantities of 

 alcohol (Klocker, 4). By this method one part of alcohol in 500,000 to 1,000,000 can 

 be detected ; 5 c.c. of the liquid to be examined are put in a tube 180 mm. long and 

 24 mm. in diameter, closed with a cork through which passes a glass tube 80 cm. long 

 and 3 mm. in diameter, the lower part not prolonged below the lower level of the 

 cork. The liquid is then boiled gently with the tube in a vertical position. If alcohol 

 be present characteristic oily drops appear in the tube ; the less alcohol present the 

 higher the level in the tube at which the drops appear. The same reaction is given 

 with acetone. 



