Nothing is added to make it rise
A traditional idli or dosa batter contains rice, split black gram, water and salt, and that’s the entire list. No yeast is bought, no culture is kept, nothing is inoculated. Left in a warm place overnight it doubles in volume, turns pleasantly sour, and becomes a different material from the one that was ground.
What does the work is a population of microorganisms already present — on the grain, on the equipment, on hands, in the air of a kitchen that has done this before. The standard account is that lactic acid bacteria dominate early and produce both the gas and the acidity, with yeasts contributing as well. This is a domesticated wild fermentation rather than a controlled one.
The two ingredients are not doing the same job
Rice supplies the starch, the bulk and the substrate — the fermentable material the microbes work on and the structure that sets when heat is applied. On its own it would ferment and produce gas, and the gas would escape immediately.
Black gram is the interesting half. Ground with water it produces a strikingly viscous, slippery batter, and that viscosity comes from proteins and complex carbohydrates that trap air mechanically during the grinding and then hold the gas produced by fermentation. It is the reason an idli is a sponge rather than a dense cake.
This is why the grinding instructions for the two are different, and why a batter ground carelessly never rises properly no matter how long it is left. Air incorporated during grinding is part of the structure, and a grinder that heats the batter while working damages both the protein and the microbes that were about to use it.
Temperature is the master control
Fermentation rate roughly tracks temperature within the range these organisms tolerate, which is why the same batter is ready in eight hours in a coastal summer and refuses to move in a cold northern winter. Nothing is wrong with the batter in the second case; the organisms are simply working slowly.
Warmth doesn’t have to be much. A closed oven with only the light on, a warm corner near a stove, a wrapped bowl — all of these lift a kitchen a few degrees and that is generally enough. Direct heat is a bad idea, because past a certain point you start killing the population rather than encouraging it.
Cold is the opposite lever and just as useful. Moving a risen batter to the fridge does not stop fermentation but slows it to a crawl, which is what lets one batch serve several days without becoming inedibly sour.
Sour and risen are two different measurements
A batter that has risen well is full of gas. A batter that’s sour has accumulated acid. These usually happen together, but they are separate processes and they can be out of step, which is the source of most confusion about when a batter is ready.
Over time acid keeps accumulating while the gas production slows and the trapped bubbles begin to escape, so an over-fermented batter is both more sour and less airy than it was at its peak. That is the flat, aggressively tangy batter that produces heavy idlis, and no amount of resting will bring it back.
Different dishes want different points on that curve. A dosa can carry, and arguably benefits from, more sourness than an idli, which is why the same batter is often used young for one and older for the other. That is a genuine tradition rather than a way of using up leftovers.
The salt argument, and what actually goes wrong
Whether salt goes in before or after fermentation is a live disagreement. Salt inhibits some organisms more than others, so it does shape which populations dominate and how fast things move, and cooks in different regions have settled on different answers with equal confidence. There is no clean verdict to report here.
The failures, on the other hand, are predictable. Too cold and nothing happens. Too long and the structure collapses. Too much water and the batter cannot hold gas, so it rises and then falls. And a batter ground so fine that no air was incorporated will underperform even when the fermentation itself went perfectly.
One more thing worth knowing: a kitchen that makes this regularly ferments faster than one that does not, because the resident microbial population builds up over time. A first batch in a new home being sluggish is entirely normal.
Common questions
My batter rose and then sank. What happened?
It passed its peak. The gas that was being produced faster than it escaped is now escaping faster than it’s being produced, and the structure holding it has weakened as acidity increased. Use it anyway, but expect denser results, and start refrigerating earlier next time.
Can I use a starter from a previous batch?
Yes, and it works well, particularly in a cold kitchen or a new one. A few spoonfuls of active batter introduce an established population and shorten the lag considerably. It will also push the flavour towards whatever that batch had become.
Why is the batter for dosa thinner than for idli?
Because they are cooked differently. An idli needs to hold its gas while it sets in steam, which requires a thicker batter with structure. A dosa has to spread into a thin film on a hot surface and dry out, and a thick batter simply will not spread.
Editor, Wah Taste
Sanaya has written about everyday cooking, street food, spice & pantry for most of the last decade and prefers a plain explanation to a clever one.
