James, I am reading with curiosity what you just said. One thing come to my mind, What if the differences of the 1794 specimen was printed in a hough flat press, compared to better technology that appeared after? Just a idea. I am curious about it too, as I just heard of it, now.

Wear and tear during a print run would show… That could be part of the explanation, I think.

. . .
Bert Vanderveen BNO

I would expect that there may have been changes in type metal composition.

IIRC, the late 18th century saw increasingly harder type metal, coupled with smoother papers that Baskerville was pioneering, which led to the kind of incredibly fine serifs and hairlines that were employed by Didot, Bodoni, et al.

It is entirely possible that there were changes in the alloys used in 20th century, which might have contributed to the changes you see. I would guess that the harder metals of late 18th century might have been perhaps too brittle to withstand pressures of early 20th century printing techniques.

It's also possible that the punches were worn since 1794 (perhaps having been struck many times). Or perhaps the revival was done by means of electrotyping, in which process some detail and refinement could have been lost.

These are just speculations on my part. James Mosley might have some more informed perspective on type metals of the two periods and casting techniques.

-- Kent.

It just occured to me that while Fry and Steele were designing types derived from Baskerville’s work, their presses might not have been nearly as good. IIRC Baskerville never profited from his work as a printer, and it seems conceivable that most printers lacked the wealth he was able to throw at his work.

I’ll probably get the librarians to photograph a few pages and ask James Mosley what he thinks.

I have this obsession about the wood flat presses. Actually, anyone with carpentry or plumbing notions would cut and put together a press. It just depends on the grade of interest and time to spare to get it. I know for fact that a press made of wood or even a cast iron flat press can not make a print as good as the late 19th or early 20th century. The platen, which presses the plate against the paper wears the type over time, and controlling the alignment of a flat piece of wood must have been a lot of fun, he he. I don't know, were they founding type by hand in 1794, or there was casters, early industrial revolution casters? James, is there type available from that time, I mean, pictures... or the real lead thing?

The type museum in London has type from that era; it’s where the Stephenson Blake collection ended up, which includes the Fry and Steele stuff. John Baskerville’s remaining equipment is at the Cambridge University press, a gift from Deberny & Peignot. And of course the Smithsonian Museum of American history has some, but their stuff is all in an asbestos-infested warehouse and won’t be accessible for a few more years, as the museum renovation is behind schedule and nothing much else will be done until then. And of course, there’s St. Bride, also in London.

I can’t believe I keep track of all this stuff. I’m such a nerd.

I don’t think that 18th-century type metal was very different from that of the early 20th, but type printed in the 18th century with hand presses on damped handmade paper, looks very different from the same type, machine printed on smooth modern papers. And although the quality of early hand casting could be very good, machine casting in the 20th century was probably better. Look at the ‘Fell types’ of the University Press at Oxford in its specimens of the 1690s and in 20th-century printing: they are cast from the same 17th-century matrices, but the later machine casting and more expert printing make a big difference.

According to the house history of Stephenson, Blake & Co., by Sidney Pollard, Fry’s Baskerville was introduced by them in 1910, not long after they had bought the foundry of Sir Charles Reed, which had all the materials of the Fry foundry, and removed it to Sheffield.

In about 1770 the Fry foundry, whose first types in the 1760s were what they called an ‘improvement’ of Baskerville’s, had also made an imitation of the smaller sizes of the Caslon Old Face types – a very close copy that is not easy to tell from the original. In 1907 Stephenson, Blake had recast this Caslon look-alike from original matrices and began to sell it under the name of Georgian Old Face. In order to add larger sizes to it they cast some types from the Fry ‘Baskerville’ matrices, then decided to add the smaller sizes of this type and market the typeface as Baskerville Old Face. At about the same date they sold sold duplicate matrices of some kind to ATF.

Incidentally Stephenson, Blake did not try to recast the original italic to the Fry Baskerville type, which is admittedly rather weedy. So, according to Mac McGrew Morris Fuller Benton made one in 1915 for their Baskerville. His model was the type by Richard Austin which was already in use in the United States and on which – some years later – English Monotype based its ‘Bell’.

Stephenson, Blake certainly had a series of original matrices for Fry’s Baskerville, but it is not easy to say how they made the revived type. There are 14 sizes of the modern type, from 72 point to 6 point. Most of them look very like the type in the Fry specimen of 1787, which I have compared them with. But one of the problems in comparing the two is that not only are the early Fry specimens more roughly printed than those of Stephenson, Blake and ATF, but all the specimens show so few characters that it is difficult to make a direct comparison between the two. One really needs to cast types from the original matrices and proof them.

And there are three sizes in the ‘revived’ Fry’s Baskerville which do not match the types in the Fry specimen at all. They are the 24, 36 and 48 point. Are they from genuine original matrices? Pollard, referring to the Georgian and Baskerville types in his history, has this rather odd passage:

‘It was a happy accident which led to the acquisition of these valuable old type faces by Stephenson, Blake & co. rather than any other modern founder, for the skill, the taste and the care required for their revival could probably not have been found in any other house. the work involved was much more than merely the casting from existing matrices, or the use of existing punches, for in the long years of neglect many of the hundreds of matrices and punches of these series had been lost, damaged or had deteriorated from various causes. the restoration of these irreplaceable models had to proceed with an eye to the exact reproductions of the effects achieved by their 18th century originators, whilst incorporating the latest improvements in technique.’

I am not sure exactly what he is trying to say, and in any case he was probably trying to summarize something that he had been told and may not have quite understood. (He was an economic historian, not a typefounder.) I think, not to put too fine a point on it, that a certain amount of highly skilled faking may have been done to achieve these ‘exact reproductions of the effects’. We know that in the later 19th century the Caslon foundry completely recut some sizes of the Caslon types, and that their claim to be casting them from matrices derived from the original punches was fraudulent.

So I suspect that the 24, 36, and 48 point sizes may possibly have been made by the resident punchcutters and matrix-makers at Sheffield in order to fill out the series. (The same thing was done quite openly in France at about the same time, in order to fill out the series of the so-called Garamont types at the Imprimerie nationale. And Hermann Zapf – equally openly – drew one additional size of Janson for Stempel.) Even the 30 point, which does appear to match the forms of one of the original Fry types (the ‘Two Lines Great Primer’), has the shortened descenders that British and American typefounders put in most of their types in order to adapt them to the standard ‘lining system’ which gave all types of each size a standard baseline. So you could set Fry’s Baskerville and Cheltenham in the same line, supposing you wanted to.

Please excuse this long post, but it seems useful to get some of this story on record. The original Fry matrices do exist, among the Stephenson, Blake materials at the Type Museum, but since its collections are inaccessible at the moment it seems unlikely that anyone will be able to get at them for some time.

Some detailed records were made when the matrices came to Sheffield, linking them with early type specimens and giving their location in the foundry. I looked at them when I visited Stephenson, Blake in Sheffield in August 1996, just before it gave up casting type. And I made some photographs with a little camera that was not ideal for the job, but here are the matrices for what appears to be the largest size of capitals, the Five Lines Pica or 72 point. Note that there are two alternative versions of Q, both of them good. How else would one know they existed?

James, thank you very much for your help on this topic, both here and via email. You have brought to light a great deal of detail about a typeface that is usually given a few sentences at best. I am trying to get new images of the 1766 and 1768 specimens from the libraries in Providence and Stockholm, I will report back on how that goes.

Thank You