The Romans had pumps, aqueducts, water wheels (which were used to grind flour, saw wood & stone etc), valves & pipes (they are known to have used stone, clay, lead & copper for pipes).

All the ingredients in fact for a pneumatic tube system for delivery of small packages.

So, would an alternate past where the Romans had actually put all these things together to produce a pneumatic tube postal service for the city of Rome be plausible?

The air pumps would be powered by water wheels.

Only for government buildings and districts occupied by wealthy citizens of course.

How the Romans made lead pipes

Copper pipes (if used) can be made the same way (hammered around an iron or steel mandrel) so any issue of standardization is more or less inherently solved by the production method.

Message-tubes to feed into the pipes can also be made this way hammered around a slightly narrower mandrel with leather or felt (rather than rubber) washers to give a relatively airtight fit in the pipes.

  • 2
    $\begingroup$ I doubt it would be realistic or plausible. Thats a lot of moving parts, pressure and precision needed to create a partial vacuum and capsule system. $\endgroup$ – Shadowzee Jan 31 at 1:43
  • $\begingroup$ @L.Dutch : good pick for the two spare tags, thanks :) $\endgroup$ – Pelinore Jan 31 at 2:10
  • 3
    $\begingroup$ Why don't they just put the packages in crisscrossing aqueducts? $\endgroup$ – svenvo7 Jan 31 at 4:48

Reality check, sadly, I'd have to say is:


The only thing truly implausible about the scenario is slaves. Rome had a lot of them. It becomes a simple matter of economics: it's much cheaper to build an aqueduct than it is to supply Rome with water via lines of slaves forming a bucket chain all the way up into the mountains; whereas it's much cheaper to tell a slave to "pop on over to M Lucius Cicero's office with this packet of letters" than it is to build a pneumatic tube system on a large scale.

Is it technologically possible? Absolutely! Does it satisfy the cool factor? Sure! But is it plausible in a heavily slave dependent Roman Empire? That I'd have to say is no.

  • 5
    $\begingroup$ "But is it plausible in a heavily slave dependent Roman Empire? That I'd have to say is no." then I'll just have to build in a sudden shortage of slaves resulting from a loss of large portions of the empire :) $\endgroup$ – Pelinore Jan 31 at 3:37
  • 4
    $\begingroup$ @Pelinore -- alternatively, you could just ignore me and do it anyway! (That's what I would do in your situation!) Check out S.P. Somtow's Aquiliad for an alternate Rome hooked on steam. Pneumatic tubes would fit right right in to that kind of scenario! $\endgroup$ – elemtilas Jan 31 at 3:39
  • 3
    $\begingroup$ One reason to do it which counters the slave issue is security. There's a reason shops use vacuum systems to transport money within the shop. (Even modern supermarkets do it.) If you're worried about your slaves getting robbed, or if you're transporting secrets which you don't want to trust a slave with, then a secret pneumatic tube system keeps whatever you're transporting away from prying eyes. Unless someone can break into the tube partway along, of course, but that maybe gives you a plot point. $\endgroup$ – Graham Jan 31 at 10:52
  • 1
    $\begingroup$ @Pelinore if there were such a shortage of slaves, they would have an extremely long list of far greater concerns (like food) than trying to develop a pneumatic tube message system. The benefits simply are not anywhere near the cost, especially considering they would also be lacking the slaves to build the system too. What exactly is the problem being solved, and why is this exceedingly expensive and truly experimental system the solution? $\endgroup$ – pluckedkiwi Jan 31 at 13:48
  • 1
    $\begingroup$ I think all things considered, it's really not all that expensive --- Rome already had a robust pipe making industry. They were terribly good at plumbing after all! The problem being solved, of course, is the improved efficiency of bureaucratic communications. $\endgroup$ – elemtilas Jan 31 at 13:52

I'm voting for Yes

Lead pipes can be hermetically sealed. It's lead. It could be hermetically sealed with a candle (well... not a candle... but if they can extrude lead pipe they can obviously melt it).

It's possible that the Romans had springs. This is good! Because the manufacturing tolerances of the pipe probably wouldn't allow even a loose-fitting canister. BUT! Throw some spring-loaded legs with wheels on the thing such that a smaller canister can roll along with shock absorption... now we're cooking with gas!

But, those springs have consequences. No vacuum. But... that could easily be fixed with a leather collar to seal around the canister. More complex than I like... but it's solving problems.

I'm not super fond of pneumatics for this purpose. I don't believe sealing the pipe is the problem. I do believe creating the vacuum in the first place could be. Pumps are good for moving fluids, but you need really, really, really well sealed pumps to create even a mediocre vacuum.

Water, on the other hand.... Fill the pipe with water from the sending end. When the canister arrives, the recipient closes a ball valve, allowing the water to drain on both sides. How does the sender know when to let the water drain? When the pressure equalizes.

Now, officially, measuring pressure wasn't something that happend until the 1500s. But it wouldn't be hard to have a small pipe affixed to the main line, probably pointing up, such that when the line pressurized a spray of water came out of it ringing a bell. That would tell the sender to turn off the water and let it drain. The line would need to be designed such that the canister didn't fall back as the water drained. Hey... you might not even need the valves I mentioned. All you need is the ability to turn water on from both sides of the pipe.

So, yes, I believe it's plausible for an alternate-history story.

  • 6
    $\begingroup$ It doesn't work with water. Water is heavy and incompressible, so in order to move the message capsule one needs to move all the water in the tube. And the pressure differential is very much harder to maintain with water than with air, because water is heavy so that differences in elevation translate into differences in pressure. $\endgroup$ – AlexP Jan 31 at 3:41
  • $\begingroup$ @AlexP, I agree it's harder with water than air, but I disagree that it doesn't work with water. I've watched sewer maintenance workers push cleaning pigs through sewer lines for hundreds of yards with just water. $\endgroup$ – JBH Jan 31 at 3:45
  • 3
    $\begingroup$ Oh, water works better than air to push things. That's not the problem. The problem is making a postal system with water. For example, consider just an imperial palace, Nero's Domus Aurea. It covered some hundreds of acres on the slopes of the Palatine, Esquiline and Capitoline hills, with a difference in elevation of perhaps 150 feet. With water, that's some 3 atmospheres pressure just in the steady state. Moreover, I cannot see how the tubes could be kept filled with water while it circulates and yet allow insertion and retrieval of capsules. $\endgroup$ – AlexP Jan 31 at 3:55
  • 4
    $\begingroup$ The pneumatic postal system of New York City was a nightmare of maintenance. Every time a jam occur you had to open a vaccum pipe, fix it, seal again then restore the vacuum. Now picture this with sealed lead pipes. The system would be in service only the days where the temple of Janus was closed. (en.wikipedia.org/wiki/Temple_of_Janus_(Roman_Forum)) $\endgroup$ – Rekesoft Jan 31 at 10:49
  • 1
    $\begingroup$ @JBH Practicality should be a key point of feasibility. What's the point of doing things that aren't going to function 9 days of every 10? $\endgroup$ – Rekesoft Feb 1 at 9:12

If you want to use pneumatic post, you need to be able to rely on vacuum. This implies that whoever manufactures the system has to ensure tight tolerances on both the pipes and the vessel, to prevent leakages disrupting the vacuum system.

I am not sure manufacturing processes in the Roman Empire could deliver a combination of narrow tolerances and large production, of the order of hundreds of meters of piping on a single building.

Moreover, lead is a poor material for this application: being really a soft metal, I doubt it would withstand the pressure differential between the vacuum on the inside and the atmosphere on the outside. You would either get breakages/deformations, or would need to use thicker walls, making again tolerances more broad.

  • 4
    $\begingroup$ I have used these systems myself & believe you're greatly overestimating the pressure differential required to move a small light package tube through the pipes. $\endgroup$ – Pelinore Jan 31 at 2:05
  • $\begingroup$ Right they are quite "loose". $\endgroup$ – Fattie Jan 31 at 21:05

Pneumatic tube diagram

It is entirely possible to have a series of pneumatic tubes working in ancient Rome. The technology did not come into play until the early 1800's, but there is nothing that could have stopped it from happening at an earlier time.

Romans had lead pipes. They had waterways to transport water to homes. By adding leather and grease gaskets to connect the pipes, you can send objects down the pipe. You could use the existing aqueducts to run the pipes. The issue is having a standard circumference, say 6", a cylinder with leather gaskets on each end and a source of air pressure. This isn't much different than the version first operating in London in 1850. The air pressure on modern systems is only 15 psi. You could use a windmill, a water wheel to provide blower power. A blower is a simple squirrel cage fan turning fast enough to push air. Your bathroom fan is powerful enough to do the job.

Good luck with your series of tubes.

  • $\begingroup$ "The air pressure on modern systems is only 15 psi" : thanks, looked for that myself but couldn't find it :) $\endgroup$ – Pelinore Jan 31 at 4:06
  • 1
    $\begingroup$ The air pressure on modern systems can be only 15 PSI because we've got tight manufacturing tolerances. In an ancient Roman system, you're going to need to generate much higher pressure at the blower to compensate for leakage, which in turn means burst seams from time to time. $\endgroup$ – Mark Jan 31 at 22:29
  • 1
    $\begingroup$ @Mark You don't need much air pressure to send a spit wad through a straw. So long as most of the gap in the tube is filled with a leather gasket, a letter in a container will definitely propel forward. I feel as if you could drop the pressure even lower because the system does not have to operate like a modern system. $\endgroup$ – gwally Feb 1 at 0:11

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.