56 C. W. M. Poynter 
ducts persist above, their union occurring opposite the third or 
fourth vertebra, see my case, page 34, illustration fig. 25, and 
examples below. 
There is a specimen in the Warren Museum in which the duct 
passes up on the right of the aorta to the third dorsal vertebra 
where it crosses to the left, divides into two trunks and these con- 
tinue as separate channels to their termination, one in the venous 
angle and the other in the jugular behind; both are left ducts. 
All the arteries are normal. 
Persistence of both the right and left duct is seen in conjunc- 
tion with double aortic arch, Watson (1877), and in low origin 
of the right subclavian, my case, page 34; it may be noted in the 
latter that the left duct is smaller than the right, suggesting that 
this is an intermediate stage in the cases of persistent right duct 
only. 
Persistence of the right duct and obliteration of the left has 
been observed in cases of right aortic arch, of the type presenting 
a left innominate, Thomson (1863), Reid (1914), case Il. It 
has also been observed in right aortic arch of the type in which 
the left subclavian is the last branch, Combes & Christopherson 
(1884). It may be found in cases of low right subclavian, see 
below, although in these double duct is almost as frequently en- 
countered. 
From the above it will be seen how difficult it is to discover a 
relationship between arterial variations and thoracic duct varia- 
tions. It would seem that the factors which operate to produce 
a low right subclavian artery also tend to establish the right duct 
as a functioning structure. If as pointed out, page 33, the low 
right subclavian is related to the cases of right aortic arch of the 
type having a low left subclavian we would expect, reasoning as 
above, that the left duct would be the one developed and this is 
the case. This result in the case of the right arch with low left 
subclavian, however, does not prove that the disturbing develop- 
mental factors have influenced the duct for we are here dealing 
with a duct that is normal. 
I would conclude from a consideration of all of the cases of 
anomalous thoracic duct, taken in conjunction with arterial varia- 
284 
