The great bottle lid debate
25 February 2026
Some scientific debates are profound. Determinism or indeterminism. The interpretation of quantum mechanics. The origin of life. And then there are the debates that follow you across your entire career, resurfacing reliably every few years like an old injury in cold weather. For me, as a convinced genetecist, such debates can be about things as profane as bottle lids.
Let's be specific: when you open a bottle at the bench, does the lid go face up or face down?
I first encountered this question as a volunteer in a chemistry lab. The answer there was unambiguous: face up, always. The bench is a landscape of solvent residues, reagent splashes, and the accumulated chemical history of every experiment conducted in the vicinity. Believe you me, we did experiments with weighing in phenolphthalein indicator powder and then spray the area with sodium hydroxide solution to see how far the particles spread. The results of such experiments are spectacular! So, place the sterile inner surface of a lid face down on that bench and you have introduced deterministic chemical contamination. The bench, in other words, is the enemy.
Then I started working with microbiologists.
In microbiology, the answer is equally unambiguous, and precisely opposite: face down, always. Air contains particles. Particles fall. Microbes land on surfaces and grow. Leave the sterile inner surface of a lid facing upward and you've essentially invited every airborne microorganism in the room to join your experiment uninvited. The air, in other words, is the enemy.
Both positions are entirely logical. Both are correct. Both are wrong. They are simply optimising for different threats. The chemist fears the bench. The microbiologist fears the sky. Neither has bothered to mention this to the other, which is how a perfectly reasonable disagreement becomes a decades-long source of mutual bewilderment.
As a side note: sometimes you have to foster your prejudices. I was trained by a chemist, and I worked for quite some time in an operating theatre of a hospital. Surfaces are evil, and things touching surfaces, the floor in particular, are instantly thrown away. The bottle lid goes up. Always. End of discussion.
Flaming the rim: elegant in theory
But the bottle lid debate is merely the opening act. The real theatre begins with aseptic technique.
Flaming a bottle neck before use has two justifications. The first is straightforward: burning off contaminants on the rim. Even though how these contaminations should end up in the bottle is not self-explanatory to me. Still, sensible, low-cost and somewhat hard to argue with (believe me, I tried!). The second explanation is more elaborate. Heating the mouth of the bottle warms the air inside, causing it to expand and flow outward, theoretically preventing contaminated room air from entering. The expanding sterile air acts as a kind of invisible bouncer at the door.
This is thermodynamically accurate. Briefly. But the moment you remove the bottle from the flame, the air cools, contracts, and pulls inward. The bouncer, it turns out, only works while you are watching. In practice, the real protective factors are speed, correct angling and not waving the tube around as if conducting an orchestra. The expanding air story is a useful teaching simplification, but it is a simplification nonetheless.
Does flaming work? Perhaps. The mechanistic explanation may be somewhat optimistic, but the practice persists because it is cheap and often beneficial. Sometimes that is enough.
The Bunsen burner: guardian of the bench
Then there is the Bunsen burner itself, burning faithfully at a distance that can singe the hair on your un-gloved hands (never wear gloves when operating a Bunsen! The burns the combination can cause are pure evil!). The Bunsen is supposedly generating a protective zone of sterile air above the workspace. Hot air rises, we are told, forming a shield against falling contaminants.
Hot air does indeed rise. That part is straightforward physics. But rising air must be replaced, and that replacement air comes from the sides, the very room you were hoping to exclude. The result is not a sterile dome but a narrow upward convection plume, which is considerably less dramatic than advertised.
Does the burner meaningfully reduce contamination? Perhaps, in some contexts. But it feels like the Bunsen is the next-best thing to a laminar flow hood. When this is not even remotely true.
Obviously context matters enormously here. As a geneticist I am using specialisted media, optimised for bugs such as E. coli. E. coli not only grows very rapidly, but also has a growth advantage on such medium. Sometimes a more sloppy approach is necessary in genetics experiments. For example, liquid needs to dry on the surface of plates. Will these plates be perfectly sterile if left open on the bench? No, of course not. However, does it really matter? No, because colonies by contaminants come up after 3–4 days. And by that time the plates have long gone into the bin and are possible already autoclaved. Obviously I do appreciate that in more classical microbiology experiments some contaminations can be catastrophic. Different experiments, different stakes, different levels of acceptable risk.
The actual point
None of this is an attack on microbiological technique. These practices evolved empirically, in specific environments, under specific pressures. They work, broadly speaking, and they were established long before anyone was writing fluid dynamics theses about Bunsen burners. Over time they became ritualised, and their explanations became pedagogically streamlined. Which is fine, as long as nobody starts treating the simplified version as settled physics.
What is more interesting is what happens at the boundaries between disciplines. A chemist and a microbiologist, both competent, both trained, watching each other at the bench with barely concealed alarm. What looks like carelessness from one perspective is usually just a different risk calculation. The microbiologist optimises for biological purity. The chemist optimises for chemical integrity. The geneticist, frankly, optimises for "clean enough to answer the question without throwing the experiment away."
None of these is inherently superior. They're context-dependent strategies, shaped by the specific threats each discipline has learned, sometimes painfully, to take seriously.
And then, of course, there is student work, which always fascinates me. I often do explain the bottle lid theory to my students. I highlight that both orientations have their value, depending on what you are looking for. The one thing that I ask students to do is pick one orientation – and then stick to it consistently!
It always surprises me, within seconds you can learn something quite profound about a student working in the lab. Some listen and follow your advice. A good start. Some do the opposite, consistently. Nothing wrong with that at all, at least if we are talking about bottle lid orientation. Some carefully watch the way how you are doing an experiment and, sometimes even unconcisously, follow how you are doing things. Often even without any explanations. These are the students to look out for. In my experience they make for absolutely great lab workers. Better a good copy than a bad original.
Far too often students do not pay any attention, do the opposite of what you just explained – and then do not even do things consistently. These are the majority of students and, unsurprisingly, these students are the most difficult to work with.