First, a couple of observations regarding religion and the manumission 
of slaves:

a) the terms used as labels for "servants" (including slaves) in the 
decades before the 1670s-80s in Virginia were "Christian" (meaning 
English or white), "negro" (African descent or black; usually 
non-Christian), "mulatto" (of mixed white and black parentage), and 
"Indian." The term Christian would not have been used to designate white 
servants unless most "negro" servants were non-Christian. Interestingly, 
"Christian" gives way to "white" in the 1670s-80s.

b) one can infer a likely connection between manumission and prior 
conversion to Christianity from the practices of those blacks who became 
free in the colony: e.g., the leaving of wills with religious language; 
the practice of godparenthood.

c) the connection can also be inferred from the 1667 Virginia statute 
declaring that baptism did not change the status of the person, whether 
bond or free; other passages in the statute suggest that the reason for 
passing it was 1) many masters assumed (or suspected) that baptism 
should confer freedom on a slave; and 2) they refrained, as a result 
from attempting to convert their slaves to Christianity, and this was 
deemed an unfortunate trend that needed to be reversed.

d) there is an occasional document making the connection more explicit; 
for instance, in 1648 Eastern Shore of Virginia master Stephen Charlton 
assigned a three-year-old Negro girl to Richard Vaughan to serve him 
until she reached the age of thirty, the said Vaughan  promising to 
"bringe the said child up in the feare of god..." (Northampton County 
Deeds, Wills, 1645-51, p.152). Another example is given in Warren 
Billings' collection of documents, The Old Dominion in the Seventeenth 
Century (p.169). In Lower Norfolk County in 1667, "Fernando a Negro" 
sued Captain John Warner for his freedom "pretending hee was a Christian 
and had been severallmyeares in England and therefore ought to serve noe 
longer than any other servant that came out of England...."

f) if one uses the statutes as a principal source, one would conclude 
that there were only two issues creating any significant amount of doubt 
as to the slave status of people of African descent: religious standing 
(1667 statute) and mixed parentage (white/free father, black/slave 
mother: 1662 statute). As noted above, the wished for result of the 1667 
religious ruling was an increase in the number of masters who would feel 
free to convert their slaves to Christianity; the wished for result of 
the 1662 statute was a decrease in the frequency of miscegenation.

A final point about VA/NC differences re manumission policies and 
practices. It may have been more difficult for NC owners to free their 
slaves than it was for VA owners (in the decades after the Revolution), 
but there were roughly the same numbers of free blacks (proportionally) 
in these two states in 1790 and 1810. The census figures presented by 
Ira Berlin in Many Thousands Gone (p.372) indicate that 5% of all NC 
blacks were free in 1790 and 6% in 1810, while the comparable VA 
percentages were 4% in 1790 and 7% in 1810. In absolute terms, the 
number of NC free blacks had increased in that 20-year period by a 
factor of a little more than 2, while VA free blacks had increased by a 
factor of around 2.5. The difference is of some significance, of course, 
but is not quite as extreme as suggested by Paul Heinegg in his posting.

Doug Deal
History/SUNY Oswego