Gulping like a fish is not very attractive, but for a fish it's an essential part of everyday life. Fish ‘breathe’ by swallowing water and pushing it through the respiratory chamber and over the gills. The gills themselves are made up of a large number of filaments that fill the respiratory chamber, and house a complex network of blood vessels that allow efficient uptake of oxygen. The question is, how does water weave its way through the gills, and are the gills strong enough to resist the resulting pressure? In the early 1960s and 1970s a lot of work went into understanding how much hydrodynamic resistance gills put up as water flowed over them. However, often the experimental and the predicted values didn't add up, explains James Strother, currently a post-doc from the Janelia Farm Research Campus, USA, ‘You have these two, what should be, trusted sources of information in conflict, so I thought there's a reason to suspect there's more to the story, an interesting phenomenon that we haven't observed before.’ So during his PhD in Matthew McHenry's lab at the University of California, Irvine, USA, he decided to investigate further (p. 2595).

Strother started his investigation by carefully removing a portion of the gills and placing it in a flow-through chamber where he could control the water flow through the gills. By measuring the pressure on either side of the gills, Strother could then work out the pressure over the gills. At flow rates lower than 65 ml min⁻¹ pressure across the gills increased linearly as expected – the faster the water flows, the more pressure there is acting on the filaments.

From these measurements Strother then calculated the resistance the filaments were providing and compared this with a value he had predicted using morphological measurements of the gill filaments. He was astonished – the values were still very different, with the measured resistance nearly four times greater than predicted. He explains his surprise: ‘the experiment was designed so that all of the complicating factors [such as active movement of the gills] that could obscure the relationship were removed. So this is a case where the measured and predicted value should match up very closely, but even in this situation there's still this difference.’

However, this wasn't the only interesting observation in store for Strother. At flow rates above 65 ml min⁻¹ pressures across the gills started to plateau, indicating that the gills were providing less resistance. When he seeded the water with tiny diamond microparticles, which glinted when illuminated, he was able to film the flow patterns over the gill filaments. At low flow rates, all the water flowed uniformly over the gills, ensuring maximal gill–water contact. However, at flow rates above 60 ml min⁻¹ things started to change, and by the time flows were ramped up to 80 ml min⁻¹ flow patterns were significantly altered – the higher pressures caused the tips of the filaments to deform and allow large amounts of water to shunt right past them. Fish will increase their flow rates when they're exercising or in hypoxic situations; unfortunately, however, water flowing through the gaps is wasted effort, as it never contacts the gills. Strother observed that this became even worse at the higher flow rates above 150 ml min⁻¹, where vortices developed, hindering the passage of further water.

While Strother hasn't quite yet figured out why predicted and measured values of hydrodynamic resistance still differ, his work has highlighted how biomechanics can affect how effective the gills are in taking up oxygen.

• © 2013. Published by The Company of Biologists Ltd