168 
of a siphon, without disturbing the residue, which is then transferred to. 
the filter, washed with boiling distilled water, dried and weighed. The 
loss denotes the amount of starch, and the residue consists of albuminoids 
and cellulose. These may be separated by treatment with water con- 
taining 5 per cent. of caustic alkali, waich will remove the former, leaving 
the latter as a residue. 
The ash is determived by taking another sample of 2.5 to 3 grammes, 
and incinerating it ina platinum crucible. The mineral matter is then de- 
ducted from the various principles with which it may have been removed 
in the process of extraction. We have determined by experiment that 
extraction of the different substances removes the following percentages 
of the whole amount of mineral matter present for each of the constita- 
ents. Thus the amount removed with— 
Per cent. 
Siti. 5 625 2255 Se eee eee 2 ee Soe 7 a 
TAD nGOriR and COLULOSO- 0-2. < bacic. cas - 62 Soe ne tose ne sete cece ese see 10.5 
EEE er Sioksen is co Sa cw cided sade sales oble S EER eens chee: cos cats tics eee eee 37.6 
CEILS ee yo Se as vid caleme bebe ceiocie es = = epee oe oe pases cite oe cides ae eee 44,4 
The nitrogen should be determined asa check upon the zein and 
albuminoids. ; 
Since a great deal of time must necessarily be lost in working the 
above method, we prefer the following, which, although it involves the , 
same principles, may be prosecuted in much less time. 
For this method weigh out five samples of 2.5 to 3 grammes each, 
place them upon porous cones which have been previously washed and 
ignited, and numbered respectively 1, 2, 3, 4, and 5, heat at 130° C, until 
constant weight is obtained. Treat the contents of cone No. 1 with 
bisulphide of carbon, dry and weigh. Before weighing, the residues 
must be well dried at 130° C. Loss in weight of cone and contents = 
moisture and oil. 
Treat the contents of cone No. 2 with bisulphide of carbon and alco- 
hol of 85 per cent. consecutively, without removing the cone from the 
filteringfunnel. After thisis done dry and weigh. Loss— moisture, oil, 
zein, and sugar. In the same manner treat the contents of cone No. 3 
with bisulphide of carbon and water. Dry and weigh. Loss = moisture, 
oil, gum, and sugar. Treat cone No. 4 with bisulphide of carbon, alcohol 
of 85 per cent., and water. Loss = moisture, oil, gum, sugar, and zein. 
Treatcone No. 5 in the same manneras No.4, and, without drying, transfer 
the contents of the cone to the tall beaker described in the former 
method, and proceed with treatment for starch as described above. 
After the starch is removed, the albuminoids, cellulose, and mineral 
matter are determined as in the preceding method. 
We give below the composition of two varieties of Indian corn, which 
were analyzed according to the above method. These were chosen to 
represent the crops of the North and South; as was stated before, the 
yellow corn represents the Northern crops, aud the white corn those of the 
South. On comparing the two analyses it will be seen that the variation 
in composition is not very great, the gum, sugar, and albuminoids being 
the most variable. 
No. 1 is 4 sample of yellow corn grown upon the farm of the Commis- 
sioner of Agriculture, in Pennsylvania. Although the analysis of this 
sample was published last month we reproduce it here for c-mparison, 
and give it caleulated for dry substance. No. 2 is a sample of white 
corn grown upon the eastern shore of Maryland. 
