In this article it is proposed that The avian respiratory system has a two-way air-flow which ventilates the lungs twice in each respiratory cycle. This is an alternative view and is contrary to the commonly accepted explanation of how birds breathe.
The avian respiratory system has two main functions: first, gas exchange allowing oxygen in and carbon-dioxide out; second, thermal regulation by air cooling. To maximise the gas exchange between the air and the blood, the partial pressure difference needs to be maintained. If the air and blood are relatively stagnant, the oxygen partial pressure equalises quickly and the gas exchange slows. In bird lungs that does not happen because the air in the lungs is kept moving and also a counter-flow or cross-flow system has evolved in which the air moves in the opposite or crosswise direction to the blood flow so that oxygen-rich air constantly flows past relatively oxygen-depleted blood.
The cooling function of the system has to deal with three main circumstances: first, resting in cool conditions; second resting in hot conditions and third, exercising ie flying in cool conditions (presuming that birds do not over-heat while flying?). Clearly, increased air-flow in the respiratory system will promote oxygenation and cooling, which can be achieved with a faster or deeper breathing rate. However, when over-heating while resting, an increase in air-flow could lead to over-depletion of carbon-dioxide leading to an acid/base imbalance in the blood (alkalosis). Birds deal with over-heating by panting; rapid, short breaths move a small volume of air back and forth through the respiratory system causing heat to be ejected without a large amount of gas exchange.
The two lungs are dense and packed close up to the dorsal ribcage. The trachea subdivides into two bronchi, each supplying a lung. Within the lung, the bronchus subdivides into several secondary bronchi and these are inter-connected by tertiary bronchi or bronchioles. All of these air-vessels are open-ended so that air flows through them during the respiratory cycle. Branching off the bronchioles are short, blind-ended air-capillaries where the gas exchange takes place. The tertiary bronchi are gathered into secondary bronchi which emerge from the lungs into the air-sacs. The primary bronchus becomes progressively narrower as it subdivides into secondary bronchi, while the secondary bronchi become progressively wider as they gather the tertiary bronchi proximal to the air-sacs. The lungs do not expand and contract during the respiratory cycle and birds do not have a diaphragm. Instead the airflow through the lungs is caused by the expansion and contraction of the air-sacs caused by the movement of the ribcage and sternum.
There are normally up to nine air-sacs, one of which, the inter-clavicular air-sac connects to both lungs. The air sacs are thin-walled bladders which act as bellows to promote the flow of air through the lungs; they do not have a copious blood supply and do not take part in gas-exchange. They are tucked among the viscera and connected with some pneumatised bones to help with cooling the musculature and the internal organs.
The path taken by the air through the lungs is a subject which is contentious. There is research which has shown that in some air-vessels the air flows more in one direction that another (Bretz and Schmidt-Nielsen, 1970). Also, gas analysis suggests that the air in the posterior air-sacs is more oxygen-rich while the air in the anterior air-sacs is more carbon-dioxide-rich. In most illustrations, the primary bronchus is shown connected directly to the posterior air-sacs while in some diagrams it is shown to by-pass the lungs altogether. All of this has led to the explanation that air flows uni-directionally through the lungs and requires two respiratory cycles to achieve this. But how can avian respiration be super-efficient if it takes two respiratory cycles to ventilate the lungs once in a one-way system?
There is no known valve system in bird lungs which would support a uni-directional flow. Also, if all the air-sacs inflate at the same time and deflate at the same time, under the influence of the expanding and contracting rib-cage and the lungs do not expand and contract, then all of the air-sacs will be at the same pressure at any particular time. There is no pressure gradient which would drive air from one air-sac to another; the pressure gradient must be between the bronchus and the all of the air-sacs acting collectively. It is the case that all of the air-sacs are effectively connected by secondary bronchi as part of a network within the lung. However, that does not mean that air flows from one air-sac to another. That cannot happen because the respiratory cycle is powered by expansion and contraction of the rib-cage, associated to some extent with the flapping cycle and therefore, all air-sacs are at the same pressure at any particular time, which is illustrated in Ornithology by Gill 2007 after Boggs et al 1997. It may well be that in some of the airways, the air flows more in one direction than another; however, that does not mean that there is an overall uni-directional flow. Although there is a direct and shortest path from the primary bronchus to the posterior air-sac, that does not mean that all of the air takes this path. Equally, just because there is a shortest path between air-sacs via secondary bronchi, does not mean that air flows from one air-sac to another.
A two-way flow of air in the lungs
As the rib-cage expands during inhalation, all of the air-sacs expand at the same time and air is drawn-in through the trachea and the bronchus, through the lung and into the air-sacs. During exhalation, the ribcage contracts, the air-sacs contract and air is forced out through the lung, through the bronchus and out of the trachea. This means that in each respiration cycle the lungs are twice flooded by moving air and no stagnant air is left in the lungs.
Thus, it makes more sense to say that one respiratory cycle ventilates the lungs twice in a two-way system and it is the double-ventilation which makes bird-respiration more efficient. Furthermore, a two-way, pressurise / de-pressurise cycle will promote air-flow in and out of the blind-ended air-capillaries and the pneumatised bones.
If this is the case, and in the absence of any valve system, then the primary bronchus should be found to be progressively narrower as it subdivides into secondary bronchi distal to the broncus; whilst the secondary bronchi are progressively wider as they gather tertiary bronchi proximal to the air-sacs, which is indeed illustrated in Gill 2007 after Lasiewski 1972.
An animated illustration of this article can be found at this link avian respiration.