I've managed to find 13 serial numbers or partial serial numbers for the Vivitar 19/3.8 using, mainly, Google image search.

Thanks to everyone who has sent me numbers. I now have 14 complete serial numbers, 2 incomplete serial numbers (first four or five digits only), and 14 mount designations.

I find:

- All begin with 9. This is widely believed to mean that they are all manufactured by Cosina. No surprise here.

Still all 9 in the larger sample.

- The second digits are 0, 1, 2, 3, 5, 7, and 9. From such a small sample, I'd hesitate to use the missing numbers to suggest a limited range of years (e.g., 1979 to 1987 or 1977 to 1985). A larger sample will help here.

I now have second digits of 0, 1, 2, 3, 5, 6, 7, 8, 9. The only missing digit is 4, but that could just the result of a small sample.

More usefully, I've found mount designations of: "PK-A/R" (4 lenses), "N/AI-S" (5 lenses), "Y/C" (1 lens), "M/MD" (2 lenses), and "C/FD" (2 lenses). Following up a suggestion by Hank, since MD, AI-s, and PK-A were introduced in 1977, 1982, and 1983, this means that at least some of these lenses were manufactured after 1983.

Furthermore, I've not found earlier designations such as "M/MC", "N/AI", or "PK", which suggests that none of the lenses for which I have mount designations were manufactured before those dates.

Finally, this page:

https://www.camera-manual.com/vivitar-19-3.8-camera-manual-12617.php

has magazine articles from 1989, 1994, and 1996. Unfortunately, they only seem to available in hard copy, which is not very practical.

So, I think we're looking at a lens from sometime after 1983 and probably into the 1990s.

That said, I've seen no lenses for Canon EF mount, which was introduced in 1987. The flange focal distance for EF is less than that of Nikon F and Pentax K, so I'd guess the lens could have been made to be compatible. If this lens was being sold in the late 1980s and early 1990s, why was there no EF mount version? Perhaps EF was seen as a high-end mount and not a suitable target for cheap lenses like this? Has anyone seen an EF version of this lens?

- The third and fourth digits are 01, 05, 07, 08, 10, and 11. This looks like a month number rather than a week number. Again, the missing numbers might not be significant in such a small sample. A large sample will help here too.

I now have third and fourth digits of 01, 03, 05, 06, 07, 08, 10, and 11. So, 02, 04, 09, and 12 are missing. Again, the missing digits could be the result of a small sample.

- The last four digits range from 0038 to 5670. This might be a monthly index number. Twice the mean of the last four numbers (which is an estimate of the upper limit under reasonable assumptions) is 5319. So, I'd guess that Cosina manufactured about 6000 per month or about 200 per day. Is that reasonable?

The minimum and maximum are still 0038 and 5670. Twice the mean is now 5454. So, it looks like the largest number manufactured in a single month or single batch was about 6000.

If I take the second through fourth number as defining a batch, then there were at least 13 different batches. If each batch was about 6000 lenses, that's at least 78,000 lenses.

If I take the second through fourth numbers as defining a chronological order, but not necessarily a batch, then there are at least 5 cycles of serial numbers (at the end of the cycle, the second through fourth number will keep increasing but the fifth through eighth number will decrease). So, if each batch was about 6000 lenses, that's at least 30,000 lenses.

I think we're looking at tens of thousands of these lenses, but probably not hundreds of thousands.

If you have a Vivitar 19/3.8, please send me a DM with its serial number, mount type, and whether I should keep the serial number private. This will help me to make this interpretation more firm. If you are worried about sharing the full serial number, please just send me the first four digits. Thanks.

More serial numbers and mount designations would be very welcome. Thanks.

Regards,

Alan