Established in 1872, Frederick Pearce of 18 & 20 Rose Street, New York City, U.S.A. manufactured "Electrical, Mechanical and Scientific Apparatus" and it is entirely possible that the lathe was an in-house production rather than being bought in. With a swing of 6.75 inches and able to take 10.5 inches between centres the lathe was 20 inches long and looks to have been made during the early years of the 20^th century. It was a handsome machine of unusual (if not unique) design with the V and flat ways of the bed arranged not in the normal way (on the top) but instead on the front face - just like, for example, on that of machine-tool pioneer Richards Roberts (a well-preserved example is in the London Science museum), the English James Spencer, Birch, Rolls Royce, the unusual Porter-Cable and the Japanese Toyo ML1.
From its appearance, and the maker's description in the only-known surviving catalogue from circa 1908, this was a precision machine intended for high-class work. Interestingly, its size fell roughly halfway between a WW-type watchmakers' type and the common "American precision bench lathe" (as originated by Stark*) a market niche also explored in the early 1950s' by the English Company IME with their Models 100 and 300
Bed and carriage
Of seasoned cast iron, the bed was allowed to stand for several months after the roughing out, then machined and hand scraped to a surface plate.
Formed for the full length of the casting, the ways allowed the carriage to be moved right up to both ends of the bed and completely in front of the headstock - a benefit when mounting accessories or adapting the lathe for grinding or other duties. In terms of contact surface area with the bed the carriage, having two separate side-by-side locations for the slide-rest assembly, was generous and (as a quality touch) with gib strips made from bronze. The carriage was formed, in place of a conventional apron, with ways to carry a "vertical slide" and the removable compound tool rest mounted on top in a at-slot. The slide was elevated by a screw, connected at its lower end to a pair of helical gears turned by a small knurled-edge handwheel. While this complex and expensive arrangement allowed the cutting tool height to be set quickly and accurately, it also permitted better use to be made of toolpost grinding and milling attachments, a saw-bench assembly, milling conversions and other accessories.
Continued below:

Continued:
Headstock
Elegantly styled, with crisp lines, a superb finish and the maker's name engraved on the front face, the headstock echoed watch-lathe practice in having both its cone bearings and hardened spindle in tool steel. Naturally, the mating parts were lapped together to ensure a perfect fit and a long, trouble-free life at high speeds. The spindle was hollow, machined to accept draw-in standard WW collets and, with a ground thread nose, able to mount chucks and other screwed fittings.  A 4-step pulley was used, for drive by a smooth running (though limited torque capacity) round "gut" rope. Interestingly, the headstock (with its flat faces) bore a resemblance in its general appearance that used on the miniature precision Rolls-Royce lathe of the 1940s.
Screwcutting and power feed
Both the tumble-reverse and changewheel assemblies used a simple but ingenious (and possibly unique) type of mounting. Arranged so that it could rotate around the left-hand spindle bearing, the tumble-reverse bracket was carried on a boss, slotted and pierced through on a chord line with a locking screw. The two gears were mounted on studs, protruding from the outer face of a lower segment, but no separate output stud gear was used, the tumbler gears mating directly with the gear below. Changewheels were carried on a flat plate, also arranged to rotate around the bearing and set immediately inboard of the tumble-reverse assembly. The lower section of the plate was slotted, to allow for adjustment, but as only one gear-mounting stud was provided this would have placed a severe restriction on the range of pitches available nor allowed the setting of a compound gear train for a carriage fine feed. Unfortunately, in addition, there were serious limitations to the employment of screwcutting and the power sliding feed. The first was a lack of backgear, and hence no gear-driven slow speeds - a problem that could have been partially alleviated by using a suitable countershaft to obtain  125 r.p.m., or thereabouts. However, even if this problem could have been overcome, there remained the difficulty of a serious design fault - a full nut on the fine-thread, 20 t.p.i. leadscrew combined with the lack of a dog clutch (or other means) of instantly stopping the drive. Once the changewheels were engaged, and the carriage moving, there was, apart from stopping the motor or using a fast-and-loose pulley assembly on the countershaft, no means of releasing the cut - and certainly no method of stopping it accurately at the end of a thread. The only redeeming factor was the use of a very fine pitch, accurately-made leadscrew (within 0.003" error in 12 inches) for this, combined with a large handle at the tailstock end of the bed, allowed the operator to maintain a steady rate of slow feed by hand. Sadly, the opportunity was not taken to fit the handle with a micrometer dial to read the position of the carriage accurately.
Continued Below:

Continued:
Compound slide rest
If the screwcutting arrangements disappointed, the slide rest would surely have delighted. Quickly detachable, in the manner of a Rivett 608, the unit was self-contained, with good-sized balanced ball handles, and could be adjusted laterally on top of the vertical side. It was locked in place by a long T-bolt, with a wing nut placed conveniently at the bottom of a long tube. Although unprotected against swarf and dirt the cross-slide screw was fitted with (for the time) a remarkably large micrometer dial, graduated in 0.001" intervals and locked by an outboard knurled ring that pushed the dial inwards without disturbing the setting. Even the 360° swivel top slide was equipped with a feed-screw dial of decent size - when most contemporary machines were not - and also featured a particularly large and clear set of degree markings. Simple, but surprisingly effective and adaptable, the toolholder was just a triangular clamp with a levelling screw at the apex.
Tailstock
Made in one piece - it could not be set over on its base to turn slight tapers - the tailstock carried a spindle bored to take the same collet as the headstock, centres being the solid-collet type. Although the lack of a sole plate for set-over might not have mattered when the lathe was new  - everything being scraped into perfect alignment - in the machine's twilight years, as the base wore, having somewhere to insert shims to restore centre height alignment would have been useful. Of a type used on other precision lathes, a slit in the casting closed down by a screw, the method of locking the spindle was crude and would have been a nuisance when the parts were old and poorly fitting.
The Pearce is very rare and, should you have one, the writer would be very interested to hear from you.
*Levin, Bottum, American Watch Tool Company, B.C.Ames, Bottum, Hjorth, Potter, Pratt & Whitney, Rivett, Wade, Waltham Machine Works, Wade,  Pratt & Whitney, Rivett, Cataract, Hardinge, Elgin, Remington, Sloan & Chace, W.H.Nichols and (though now very rare) Ballou & Whitcombe, , Sawyer Watch Tool Co., Engineering Appliances, Fenn-Sadler, "Cosa Corporation of New York" and UND.

Both the tumble-reverse and changewheel assemblies used a simple but ingenious (and possibly unique) type of mounting with each pivoting on a boss formed around the left-hand headstock bearing.