e 
8 PROCEEDINGS OF THE NATIONAL MUSEUM. VOL. 62. 
led to a preliminary optical examination which showed that the acicu- 
lar crystals were not natrolite but were more probably tavistockite 
or wavellite. More detailed optical measurements made by Dr. E. 
S. Larsen indicated that, in all probability, the mineral was wavel- 
lite. The properties are as follows: Biaxial, positive (+), 2V 
medium large, (460°). Dispersion p>v so small as to be doubtful. 
Z=c. No principal optical directions perpendicular to the prism 
faces. Refractive indices a = 1.525 + 0.003, B = 1.534 + 0.003, 
y=1.545 + 0.003. Birefringence 0.020 + 0.003. There are appar- 
ently three good cleavages normal to the principal optical directions. 
In order that no question might remain regarding the identity of 
the material, such as could be spared without serious injury to the 
specimen was scraped off and examined chemically. Only 0.036 
gram of pure material was available. This gave a loss on ignition 
of 30.09 per cent (=H,O0+F). The ignited mineral was fused with 
a large excess of sodium carbonate and the fusion leached with boil- 
ing water. The residue of insoluble material consisted of 0.0022 gram 
of Fe,0,, possibly largely derived from reagents. It contained no 
lime. The filtrate gave abundant reactions for alumina and phos- 
phoric acid. 
Although the crystals were very minute, averaging perhaps } mm. 
in length and one-fifth to one-tenth this in diameter, it was found 
possible, after several trials, to secure one which gave satisfactory 
measurements on the 2-circle goniometer. Since, as pointed out by 
Wherry,°® the axial ratios of wavellite are probably variable and are 
not accurately known beyond the third decimal place, the angles 
measured were used to calculate axial values for the best crystal 
measured from the Arkansas specimen. The measurements gave as 
follows: 
Measurements. Difference. Average. 
Form. | 
¢ p ¢ p ¢ p 
° / fe} / re) / ° {2 ° / ° / | 
p(101)| 89 55 | 36 10 
90 03 36 10 0 08 0 00 89 59 36 «10 
gives p,=0.7310 
s Gr) 60 26 40 06 
60 15 | 40 18 | 
GOpe Seis he 20s lee fe ean 
60 40 40 O1 | 
gives p,=0.7344 | 
0 17 | 60 28%| 40 10 | 
The values for p, derived from the two forms are not in satisfactory 
agreement; and since the faces of neither form gave conspicuously 
6Edgar T. Wherry. Notes on mimetite, thaumasite, and wavellite. Proc. U.S. Nat. Mus., vol. 54, 
pp. 373-381, 1918. 
