146 
Journal of Agricultural Research 
Vol. XXVIII, No. 2 
well as Straus (Table IV), also give weights for larvae of known age, but 
in order to eliminate the danger of variations due to the effect of different 
seasonal and environmental conditions, the average age of the larvae analyzed 
from various groups of 25 was determined by comparison with a series of weigh¬ 
ings of larvae of known age that were made during this same period {35 ). The 
various series of weights, with the corresponding determinations of reducing 
sugar, were arranged in age groups, 24 hours apart, as shown in Table V. In 
some cases, such as the small two-day larvae, or the quiescent prepupae, where 
the amount of unassimilated sugar is small, 50 larvae were taken for analysis, 
but usually 25 proved satisfactory. 
PREPARATION OF MATERIAL FOR ANALYSIS 
Several difficulties were encountered in the preparation of material for sugar 
determination. At first, attempts to extract the sugar were made by macerating 
the larvae with distilled water and filtering through filter paper. This produced 
a cloudy opalescent liquid, indicating the presence of colloidal material, and this 
solution did not give the characteristic reaction with the Benedict reagent. 
Various clarification methods were tried. Precipitation with both neutral and 
basic lead acetate {10 y p. 276) solutions proved unsatisfactory, something still 
remaining to interfere with the reaction. Mercuric nitrate solution, which is 
sometimes used to clarify liquids of animal origin such as blood, urine, and milk, 
was tried {10 , p. 447). This method occasionally gave good results, mainly 
with the younger larvse, but often with older larvse and prepupae the colloidlike 
material still remained in the filtrate, interfering with the reaction. Furthermore, 
because of the numerous filtrations necessary to remove successive precipitates, 
it was feared that more or less sugar is lost by adsorption to those precipitates, 
even with careful washing. An attempt was made to clarify by filtration with 
suction through a celloidin membrane, and this gave a clear solution which reacted 
well with the test solution, but the method required too great time. The method 
finally adopted was by extraction with 50 per cent alcohol, similar to the method 
used in the extraction of sugars from grains and similar products {11). This 
method proved successful, since the alcohol causes precipitation of all solid 
matter, giving a clear filtrate which reacted properly with the Benedicts reagent. 
Since glycogen in water solution is colloidal in nature, and thereby difficult to 
remove by filtration from such a solution, it is doubtless the glycogen present in 
the larva which prevented clarification and interfered with the reaction. It is 
possible for this reason that Straus {43) failed to demonstrate reducing sugars. 
To determine this point, a small amount of glycogen was added to a known solu¬ 
tion of dextrose and tested with the copper sulphate solution, and the known 
reducing sugars could not now be demonstrated quantitatively. Since glycogen 
is insoluble in alcohol {10 , p. 44$) the 50 per cent alcohol precipitates the glycogen 
and thereby removes materials interfering with the reaction in the filtrate. 
Even though there may still be a small loss of reducing sugar by adsorption or by 
some other means, the results obtained are of value for purposes of comparison. 
If any reducing sugar is lost by the method adopted, the amount is exceedingly 
small and may therefore be disregarded, since repeated washings failed to demon¬ 
strate its presence. 
TECHNIC ADOPTED 
After weighing, the larvse are removed to a small porcelain mortar and mace¬ 
rated in 30 cc. of 50 per cent alcohol. This material is then washed carefully 
into a small flask and allowed to stand from two to three hours before filtering. 
The precipitate is washed with 50 per cent alcohol. The filtrate is then made 
up to 50 cc. with distilled water, and run into a burette. Five cc. of the stand- 
