If it's true that military research played a big part, e.g., in the development of the internet, then this is also valid for a number of other fields as well. Indeed, it was not only road-vehicle-, but also and particularly, aviation engines, which interested them. The latter had really pushed the progress of the combustion engine, unfortunately, their aims and efforts were always targeted at winning a war. The way that BMW, a relatively new player in the aviation engine field, towards the end of the first world war, tried to trump the well established Daimler product, is indeed, exciting reading. They even set a new altitude world record, unfortunately by then it was too late.
The quest for altitude brought new insights into the field of carburetors, after all, the altitude and thus also the air-pressure had to be respectively changeable. Also radial- or rotary engines were interesting for aircraft. For the first time an odd number of cylinders was used in the in-line engine. In the construction of vehicles there are often still two cylinders cast as a pair, also with engines having four and more cylinders. The one-piece engine blocks with up to 12 cylinders or even 2x8 cyinders only asserted themselves slowly. Mostly they were mounted, filling the frame completely and can be directly bolted together.
It was not a simple matter to use them either, actually, a chauffeur was necessary. Even when crank-starting the smaller models, one could end up with a broken arm. Right at the beginning, there was not even an oil pan, a hand-pump had to be regularly operated, and after lubricating, the oil landed on the road. The combustion engine only slowly improved, e.g., with the appearance of immersion- or pressure circulation lubrication. There was also much less oil residue in the exhaust gases. In the USA there were already electric starters. In america, driving oneself was quite normal. This, together with the reduced cost of manufacturing on the assembly-line, led to widespread distribution of the motor car.
In the beginning of the 1920s, the development was divided into two sections. In Europe, even though the war had left a great deal of economic hardship, one was also considering cars for the less well-to-do public. At the same time, the motor-racing circus was flourishing. This led to very simply constructed engines on the one hand, and on the other, complex eight-cylinder engines with DOHC-technology. The simplest form was e.g., shown in the very successful four-cylinder Austin Seven with only two main bearings on the crankshaft and partly with immersion lubrication.
During the first twenty years of the last century the engine management was further developed. Until then, it was customary to have only exhaust valves which were operated by a camshaft or cam-disc. The inlet valves were conceived as check valves which only opened when the piston, on it's way to TDC, created sufficient low atmospheric pressure, not a very performance orientated process. After both valves were driven by push-rods, the two “standing valves” were accomodated in the engine block. Suitably, there was one up-draft carburetor, sometimes also more than one.
Apart from a number of other weird constructions, e.g., engines with two counter-rotating pistons in the same cylinder and two crankshafts, there was also the sleeve-valve engine invented by the American, Knight, with which Mercedes sacrificed a whole engine generation. The camshaft becomes a sort of crankshaft and moves, in each case invidually, two superimposed cylinders. With cleverly chosen openings, one could save oneself the otherwise well known valve operation. Actually, it should have made higher revs possible, indeed, the opposite was the case.