104 
POPULAR SCIENCE REYIE'W. 
on the tannin question. He says that the percentage of tannin in tea 
is very variable, and there is little doubt that this i3 in a great 
degree the cause of the erroneous estimate which English tea-drinkers 
frequently make of the dietetic value of tea. They prefer tea which 
gives a dark- coloured infusion, and has some sensible astringency, to those 
varieties which give a paler and less bitter liquor. This probably accounts, 
to some extent, for the high estimation in which some kinds of Assam are 
held for mixing purposes. The acetate of lead process seems more reliable 
for determining the percentage of tannin than the old gelatine process, and 
it is certainly easier. I have therefore adopted it. I find that a sample 
taken from a mixture of six samples of Assam tea gave 45-5 per cent, of 
tannin, while some of the highest results were — 
Per Cent. 
No. 82. Moyune young Hyson 30*0 
No. 83. Very choice Assam 33-0 
No. 1. Indian young Hyson 39 0 
No. 97. Assam tea from Dr. McNamara’s garden . . 27’ 7 
No. 75. Caper (mixed) 42-3 
GEOLOGY AND PALAEONTOLOGY. 
A New Crustacean Fossil from America . — This fossil, which was exhibited 
by Messrs. Grote and Pitt, at a recent meeting of the Buffalo Society of 
Natural Science, is of much interest. The specimen shown exhibited an 
impression of the ventral surface of a new form of crustacean allied to 
j Eurypterus Pterygotus , for which the name Ensarcus scorpionis is proposed. 
The cephalothoracic portion appears to be separate from the body, and to be 
considerably narrower in proportion than in allied forms. The legs are the 
same in number as in Furypterus. The swimming feet appear to differ by 
the straighter, less rounded outer margins. In the specimen the rhomboidal 
plates are not given. From the impressions of the joints of the swimming 
feet their relative dimension does not seem to accord with Furypterus. The 
four pair of anterior feet proceed from two elongate oral plates, of which the 
impression is very distinct. The spines of the anterior feet appear to be long 
curved, and to have an anterior direction. The absence of chelate appen- 
dages to the posterior margin of the feet is particularly noticeable. The 
first seven broad segments of the abdomen form a large ellipse. There is an 
evident and remarkable narrowing of the succeeding caudal segments. Of 
these six appear to be made out on the specimen. The surface of the cast 
is punctate with scattered triangular impressions. The cast shows a 
widening of the terminal segment, and no traces of a spiniform process are 
exhibited. This portion of the fossil was imbedded in the matrix, and seems 
to have been bent downward and outward in the specimen. The greater 
flexibility of the caudal segments may be inferred from their shape and 
position. The specimen from which the above description is drawn was 
found in the Water-lime group at Buffalo, N. Y., in the same bed which 
has furnished specimens of Furypt&'us lacustris. The length of the entire 
