Prior to 1843 - The Disruption - of which more later, anything up to 30% of birth records don't exist in the OPRs of the Established Church of Scotland due to ...
(a) the family being a member of a secession or other church, e.g. Episcopalian
(b) the record wasn't made
(c) the record was made but that OPR volume hasn't survived
It shouldn't take that long to have a look at the entries for the parishes concerned here in the First and Second Statistical Accounts where it will be found that the CoS minister responding to the Statistical Account questionnaire lists all the congregations in his parish.
Be careful, tho', as it's well recorded that folk would travel a good distance in order to attend a church according with their views, possibly not in their parish of residence.
Both on the ScotlandsPeople site - dig deep in the OPR Help!, - and in various publications the gaps in OPR records are listed in quite some detail.
Given all this, for all but the most uncommon names it's far from uncommon to find that there are various possible matches, and other evidence needs to be sought to prove that one of these is a true match.
It's one of those situations where it can be impossible to prove that there isn't a missing record that is the true match.
However attractive a possible record might be, without other evidence, it can quite frustrating frequently be the case that the hoped for link can never be proven.
There are several techniques for getting round this. One is the potentially time consuming concept of doing a one surname search in a number of parishes.
For my own surname I did this in around 30 parishes in Angus and the County of Kincardine some years ago. It took a total of 3 weeks at New Register House, spread out over a couple of months, along with considerable analysis time at home.
It was fascinating to see how trees could be built from MIs, censuses and BMDs, at least to the extent that it could be shown that there was no connection to my tree. But, as I noted, time consuming.
My own conclusion was that there was possible link from Montrose to Brechin but the only weak evidence I had was an exact naming pattern match with no such matches elswhere - but not every family stuck to these traditions; and, as always, I have no way of knowing what important records were missing.
Another method, much easier in these www days, is to trace siblings', even cousins' lines forward to modern day living relatives as the chances are good that someone out there has carried out relevant research and may just also have a key source such as a family bible, or family letters, or even other info such as occupations that make it less likely that a possible link is true.
Note also that, once you get into the mid-1850s plus/minus a couple of decades, the practice of using a family surname as a middle name, most often starting with the mother's maiden name, became much more widespread, and, for modern genealogists, incredibly useful
, not least in the statutory death indexes.
In small county parishes that 30% might be lower, but it all depends on the number of secession congregations. I'm no expert in the area concerned, but do know that in other Borders parishes to the East the Episcopalians were strong; but go look in the Statistical Accounts!
In large city parishes the number of 'missing' births can be a high as 60%, - this is based on surveys at the time.
From 1843 to 1854, after the secession forming the Free Kirk, the figure of records not to be found in the CoS OPRs can be as high as 60% even in rural parishes - again check the Third Statistical Account and other sources to see how many Free Kirk congregations there were in your parish[es] of interest.
It's also important to realise that the quality of info can vary widely, - the monumental mason misread the info given to him; the newspaper reporter is in a tearing hurry to meet a deadline, and, not least, the informant for a 1855+ death record was greatly emotionally upset so got something completely wrong; and there's also plenty of stories of wild errors in birth and marriage registrations, never mind the same factors for censuses where the 'overseas' factor needs to be added in , i.e. many census indexes are based on data input of less than ideally trained staff in companies based in the Indian sub-continent, Sri Lanka, The Phillippines, mainland China, Taiwan, etc., etc.
As regards the GROS census indexes the training given to these overseas staff is excellent, and the quality control first class, so the the transcription error rate is as low as 2 or 3%, but the same, unfortunately, cannot be said for other indexes, including one very large company that I couldn't possibly name
; but would comment that I still use their indexes since more fields in the census returns are indexed, and there are other search possibilities, e.g. given name only ......
Wullie
