210 
PHYSICS: A. A. MICHELSON 
6. The dilation of the subclavian is circumscribed, is distal to the point of 
constriction, and strikingly resembles the dilation which we have produced 
experimentally. 
7. The dilation of the artery proximal to an arterio- venous fistula and distal 
to a partially occluding band may prove to be referable to the same cause. 
8. When the lumen of the aorta is considerably constricted the systolic 
pressure may be permanently so lowered and the diastolic pressure so increased 
that the pulse pressure may be diminished by one-half. 
9. The experimentally produced dilations and the aneurysms of the sub- 
clavian artery in cases of cervical rib are probably not due to vasomotor par- 
alysis, trauma, or sudden variations in blood pressure. 
10. The abnormal, whirlpool-like play of the blood in the relatively dead 
pocket just below the site of the constriction, and the lowered pulse pressure 
may be the chief factors concerned in the production of the dilation. 
11. Bands, rolled ever so tightly, do not rupture the intima. 
12. Intimal surfaces, brought, however gently, in contact by bands or liga- 
tures do not, in our experience, unite by first intention, for the force necessary 
to occlude the artery is sufficient to cause necrosis of the arterial wall. 
13. The death of the arterial wall having been brought about by the pres- 
sure of the band, a gradual substitution of the necrotic tissue takes place, the 
new vessels penetrating it from both ends. It is, I believe, in this manner 
that an artery becomes occluded, and it is thus that a fibrous cord forms within 
the constricting band. 
r Luigi Porta. Dalle alterazioni patologiche delle arterie per la legatura e la torsione 
Milano, 1845, pp. 350, 351, plate V, figs. 3 and 5. 
ON THE CORRECTION OF OPTICAL SURFACES 
By A. A. Michelson 
Ryerson Physical Laboratory, University of Chicago 
Read before the Academy, April 23, 1918 
In a recent number of the Philosophical Magazine, an interesting method for 
correcting optical surfaces by means of the interferometer, was developed by 
Mr. Twyman. While nothing in the paper indicates that the method is 
limited to relatively small surfaces, it would appear that such an application 
to mirrors and lenses of the size of modern astronomical telescopes can hardly 
be contemplated as this would involve interferometers of at least equal 
dimensions. 
It was hoped that a modification of Mr. Twyman's method, with an inter- 
ferometer of usual size, could nevertheless be employed for large lenses or 
mirrors. 
