78 BULLETIN 846, U. S. DEPARTMENT OF AGRICULTURE. 
In the opinion of the chemists who finally passed upon the methods, 
the slight differences in results do not warrant the extra labor con- 
nected with the use of sand. : 
AMOUNT OF SAMPLE TO BE USED FOR TOTAL SOLIDS DETERMINATION. 
The reason for investigating this matter was connected primarily 
with the determinations of ether extract and of acidity of the fat 
which it was considered might be made more accurate by using larger 
amounts of sample. The question then arose as to whether these 
larger amounts of sample could be completely dried to give correct 
total solids. The results obtained are given in Table 28. 
TABLE 28.—Percentage of total solids obtained with varying amounts of sample. 
Using 2.54 grams. | Using 5.0+ grams. 
Sample 
No. 
| 
| 
Initial of analyst. 
| AS | B. A. Be 
| 
Ea cent. | Per cent. | Per cent. | Per cent. 
27 7 63 7.68 
[Bl eles hn cee ae Woe AE SNCS © 5 Ree ABU, bea OME Oe 5 27.68 7 me 
ee Ee EES Fh PE See eee fie 27.71 27.13 27.12 
APL A De ar Sapa Rt an ae gine Shae ta Eom lait eeeespened OAS © | “2732. 27.17 27.16 271A 
Ese BN SF Ni AN aa at os Or oe Re [xaes 26273 30.04 26. 86 28.10 
SESE EE ee aie eT ge A PRS Bip 27.50 27.32 27. 42 
ie. Peete. ae. SIC Pe ee 29. 58 29.77 29. 60 29.61 
Dees oe 2 URES 2 SIEIUEE, doaehe Ys gees OSE SS [+ BLT 30. 34 29.75 29. 83 
A eal Bs a nats OF MRE RRR ee eee | 29.06 28. 86 28. 92 28. 88 
Nite aye Sauce eee FS SS 2 eek oe oe are ea A | 29.09 28. 69 28.45 28.7 
WS ec Boeke ee ett he Ad oe eee Ae | . 29.97] 30.41 29. 90 30. 02 
(ee Sa Sages 30. 36 29. 92 29. 83 29.99 
Fe a peer 30. 66 30. 69 30. 84 
\E i Sees eee eee see PN Sta 30. 32 30.35 30. 44 30. 33 
Ses aL Ya eae: cae SEs 29. 46 29.51 28.55 29. 94 
Hy eg eee ee ee eee | 29.24 29. 38 28.75 22. 82 
Ay) igen See eaee ees Seee ener aor Sonne mae eee es a peat ae 29. 23 29. 06 28. 99 
He Set) a. Ei tloe Bale: | Seats ee 28. 53 28. 23 28. 41 28.31 
Bite tose Innes peat ood Sse: Sane eacee te | 29.00 28.81 28. 44 28. 65 
"Dia ae il NESS SE ee ee were ee i ees 21. 22 21.05 20. 88 20. 84 
og || HD ssoscss25e sae e sess eeeeesee ae seq 2 Sees Een Soa 19. 58 19. 47 19. 61 19.44 
SN ee ee a eee ee. eee ee ee | 20555 20. 40 20.01 20.37 
| Ss aed Speke a Ai Be lee enn SO Pale Ba ae aa pee ea oes [eee tke ay, 21.79 21.55 21.63 
| 
COS one ee ee ees eee ee | 27.58 27.31 
There is not much difference in the results obtamed by the two 
methods, but what little difference there is indicates the desirability 
of using the larger amount, as it gives more uniform results, pre- 
sumably due to decrease in the percentage error. 
