OF PISCES, BATRACHIA, AND REPTILIA. 



153 



of two interference brushes are seen in convergent light ; they are widely separated, indi- 

 cating that the axial angle is large. They are nearly symmetrical, owing to the small 

 angle of extinction. The acute bisectrix Bx a appears to be c, making the optical charac- 

 ter positive. 



These oxyhemoglobin crystals were only obtained in quantity from the blood of a 

 shad that was purchased in the market, and that had been dead some days and exposed 

 to the air. The blood was extracted from the heart and larger blood vessels in the form 

 of soft clots, oxalate was added, and the blood ether-laked and centrifugalized. Both 

 blood and preparations were full of gas bubbles. At first only the one type of crystals, 

 as described above, developed, but after some hours hexagonal plates of methemoglobin 

 began to appear in quantity, and these grew isolated or attached to the oxyhemoglobin 

 in regular growth, as described under methemoglobin below (see text figure 12). 



(2) Metoxyhemoglobin of Alosa sapidissima. 



Monoclinic: Axial ratio a :b : <! = 1.786 : 1 : 6; /?=70. 



Forms observed: Prism (110), base (001). 



Angles: On base (001) edges 110-001 A lTO-001 =58 30' on perfect and untwinned 

 crystals ; but on the larger and twinned crystals this angle runs often over 60 and aver- 

 ages about 60 40', giving a = 1.709; angle 0=edge of 110-lTO A 001=70. 



12 



'O 



a 



FIGS. 8, 9. Aloia sapidisiima Metoxyhemoglobin. FIQB. 10, 11, 12. Alosa iapidiirima Reduced Hemoglobin. 



Habit of the crystals tabular on the base (001), with short prism (110) as in the 

 shad oxyhemoglobin, the long diagonal of the plate being in the plane of symmetry as 

 above (text figures 8 and 9). The forms of twinning are exactly similar to those of the 

 oxyhemoglobin, but in the second kind of twinning (horse-type) the crystals often elongate 

 in the direction of the common prism-base edge (plate 3, fig. 15), and a very similar 

 looking twin is formed by twinning on a pyramid (111). The angle of the plates in this 

 twin (which is an interpenetrant twin) was not determined. Examples of it may be 

 seen on plate 3, fig. 15. Cleavage was readily obtained by crushing the crystals under the 

 cover, showing them to be brittle ; the cleavage is parallel to the prism (1 10) and is perfect. 



Pleochroism is strong; a pale yellow to nearly colorless, b reddish-brown, c deeper 

 reddish-brown. The spectra for light vibrating parallel to a and & were carefully observed, 

 the chief difference being in the very much stronger absorption of the blue end of the 

 spectrum, up to the green in case of 6. The absorption bands are: for a a band from 

 635 fift to 625 pp, one from 580 pp to 565 jj.fi., and one from 548 pp to 530 pp. For b the 

 bands are shifted somewhat, 640 pp to 620 pp, and very faint beyond 585 ftp, when a 

 band begins that extends to 565 pp; another in the almost absorbed spectrum runs from 

 550 up to 530 pp. The first band (635 pp to 625 pp, etc.) is the methemoglobin band 

 in the red, the other two mark the position of the two oxyhemoglobin bands somewhat 

 displaced, and the methemoglobin band in the blue-green is not visible, owing to the 

 absorption of this end of the spectrum. But it will readily be seen that this is quite 

 different from the spectrum of pure methemoglobin, which contains a band in the red 

 (640 pp to 620 pp) and one in the blue-green (513 pp to 488 pp) only. 



