Unidirectional Airflow in the Lungs of Alligators

See allHide authors and affiliations

Science  15 Jan 2010:
Vol. 327, Issue 5963, pp. 338-340
DOI: 10.1126/science.1180219

Alligator Breath

Birds have a unidirectional system of airflow within their lungs that has been attributed to the peculiarities of flight. However, Farmer and Sanders (p. 338) provide evidence that this unidirectional and more or less continuous flow of air also occurs through parts of the alligator lung; in contrast to the tidal, biphasic system in mammals. By analyzing lung and tracheal structures, the similarities of the alligator lungs were compared with those of birds. The data suggest that the unusual properties of bird lungs originated before the divergence of the alligator line from the dinosaur or avian line.


The lungs of birds move air in only one direction during both inspiration and expiration through most of the tubular gas-exchanging bronchi (parabronchi), whereas in the lungs of mammals and presumably other vertebrates, air moves tidally into and out of terminal gas-exchange structures, which are cul-de-sacs. Unidirectional flow purportedly depends on bellowslike ventilation by air sacs and may have evolved to meet the high aerobic demands of sustained flight. Here, we show that air flows unidirectionally through parabronchi in the lungs of the American alligator, an amphibious ectotherm without air sacs, which suggests that this pattern dates back to the basal archosaurs of the Triassic and may have been present in their nondinosaur descendants (phytosaurs, aetosaurs, rauisuchians, crocodylomorphs, and pterosaurs) as well as in dinosaurs.

Airflow in the avian lung is believed to be unique because gases move in the same direction during inhalation and exhalation through small tubes, the parabronchi. Although airflow is caused by volumetric changes in air sacs, the unidirectional pattern is achieved without mechanical valves. Soot-laden air was used to demonstrate unidirectional flow, and this pattern of airflow was attributed to the configuration of the bronchi giving rise to jetting and Venturi effects (in which increases in fluid velocity decrease lateral pressure) (1, 2).

Crocodilian lungs are distinct from bird lungs and are thought to have a large alveolar-arterial blood gas difference, large ventilation-perfusion inhomogeneity, and parenchyma consisting of cubicles (ediculae) (35). However, the topography of the intrapulmonary bronchus and of the first generation of bronchi is similar in birds and crocodilians (3, 6, 7).

Key features of the avian aerodynamic valve appear to be present in the alligator lung. The green bronchus shown in Fig. 1, the cervical ventral bronchus (CVB), is strikingly similar to the avian ventral bronchus that connects with the cervical air sacs. The small ostium to the CVB opens into a funnel-shaped vestibule acutely angled with the intrapulmonary bronchus, so that the bronchus makes a hairpin turn ventrocranially before coursing cranially to the apex of the lung (Fig. 1A, B). Distal to the CVB ostium, the intrapulmonary bronchus widens and becomes partially enclosed as it curves caudally and laterally and gives rise to a pair of small medial paracardiac bronchi (Fig. 1, red bronchi), a large individual dorsolateral bronchus (Fig. 1, chartreuse bronchus), and three caudal bronchi originating from a common terminal intrabronchial chamber (Fig. 1, blue bronchi). Their orifices are larger than the CVB orifice and better aligned with the intrapulmonary bronchus. Most of these latter bronchial passages spiral dorsolaterally toward the apex of the lung in a manner similar to the avian dorsal bronchi. We have discovered that the dorsal bronchi connect to each other and to the CVB through numerous anastomosing parabronchi, approximately 1 to 1.5 mm in diameter at the orifice to the bronchi (arrowheads, Fig. 1C) [supporting online material (SOM)]. As in birds, small ostia to caudoventral bronchi (arrows, Fig. 1C) occur opposite the ostia of the dorsal bronchi.

Fig. 1

Airflow in alligator lungs. Computed tomography images (left) show the hairpin turn (blue arrow) into the CVB (v) in the coronal (A) and medial sagittal (B) views. The lateral sagittal view shows the larger ostia to the dorsal bronchi (C), some of the ostia of ventral and lateral bronchi (blue arrows), a parabronchus (blue arrowheads), and the dorsal bronchus in which flow was recorded (d). The axial view (D) shows the bifurcation of the primary bronchi. An oblique dorsal view of the major bronchi is shown in (E). A simplified view shows airflow during inspiration (F) and exhalation (G) in the trachea, CVB, dorsal bronchus, and parabronchi. Hazelhoff's model of exhalation (H) and inspiration (I). ipb, intrapulmonary bronchus; le, guiding dam; ve, vestibulum; v, ventrobronchus; m, mesobronchus; p, parabronchus; d, dorsobronchus; x,y, sites of constriction.

The anatomical similarity with the avian lung led us to hypothesize that airflow might also be unidirectional in crocodilians. Previously, measurements of airflow in alligator lungs did not identify unidirectional flow (8). Perry discussed the possibility that crocodilians have unidirectional airflow, but concluded that airflow is tidal in crocodilians and that unidirectional flow evolved in coelurosaurian-grade dinosaurs (912). Furthermore, for avian-style respiration to occur in birds and nonavian dinosaurs, abdominal air sacs have been presumed to be critical (13), and the hepatic piston mechanism of ventilation of crocodilians has been presumed incompatible (14).

To test the hypothesis that airflow in alligator lungs is unidirectional, we implanted dual thermistor flowmeters in the CVB (green bronchus of Fig. 1) and a dorsal bronchus (a blue bronchus of Fig. 1) of four alligators, artificially ventilated the lungs with both negative and positive pressure inspiration, and observed that air in the CVB moved in a cranial-to-caudal direction and air in the dorsal bronchus moved in a ventrolateral to dorsomedial direction during expiration and during both types of inspiration (Fig. 2).

Fig. 2

Airflow in excised lungs. The top two traces show the direction of intrapulmonary airflow recorded from the dorsal bronchus (d in Fig. 1) and CVB (v in Fig. 1) with dual thermistor flow meters. The middle two traces show the relative amplitude of these flows. The bottom trace shows flow into and out of the trachea, recorded with a pneumotach.

To determine whether similar patterns of flow occur in vivo, we monitored airflow in the CVB during normal breathing in five alligators with single-bead thermistors and in one alligator with a dual thermistor flow meter. In the former experiments, the flow continued during the transition from inspiration to expiration rather than dropping to zero, as would be the case if the direction of the flow had reversed. In vivo recordings with the dual thermistor flow meter showed that the air moved in a cranial-to-caudal direction during both inspiration and expiration; which is the same pattern of flow observed in excised lungs (Fig. 3). The amplitude of the expiratory flow was greater than the amplitude of the inspiratory flow, as in the excised lungs.

Fig. 3

Airflow observed in vivo. The top trace shows the direction of intrapulmonary airflow recorded in the CVB (v of Fig. 1) with a dual thermistor flow meter. The small pulse of air moving toward the head after airflow in the trachea has ceased occurs when the glottis is closed and the muscles of the trunk relax (21). The middle trace shows the amplitude of the flow. The bottom trace shows flow into and out of the trachea, recorded with a pneumotach.

We also visualized flow by filling an excised lung with saline containing fluorescent microspheres. In the CVB, the microspheres flowed in a cranial-to-caudal direction as the fluid was pushed into and pulled out of the lung (movies S1 and S2). In the parabronchi, the spheres also flowed unidirectionally (movie S3). All three methods—in vivo recordings of airflow, recordings of airflow in artificially ventilated excised lungs, and visualization of water flow in artificially ventilated lungs—indicate that fluid flows unidirectionally through the lungs of alligators in a strikingly birdlike pattern.

Our data suggest that airflow in the alligator is extremely birdlike, but it is unknown how it is possible to have unidirectional flow without air sacs and with diaphragmatic breathing. A mechanism for unidirectional flow in bird lungs as a consequence of the topography of the intrapulmonary bronchi (1) that does not depend on air sacs or the mechanics of breathing is shown in Fig. 1, H and I. During exhalation, the configuration of the laterobronchial ostia may cause air to move dorsally to enter the ostia of the dorsobronchi, rather than leaving the lung directly by way of the mesobronchus (1). During inspiration, rapid flow of air along the intrapulmonary bronchus past the ostia to the ventrobronchi may reduce lateral pressure at this location because of the Venturi effect, and the low pressure may act as a suction pump to draw air past the mouths of the ventrobronchi into the mesobronchus (1). A glass model demonstrated how this geometry gives rise to unidirectional airflow (1).

The mechanism of unidirectional flow in alligator lungs is yet to be determined, but our data support Hazelhoff's model (1), in which key features of the bronchial tree give rise to unidirectional flow. During inspiration, air may jet past the obliquely oriented vestibule of the CVB to enter the larger dorsal bronchial openings and reduce lateral pressure at the CVB orifice to draw air from the CVB into the intrapulmonary bronchus. During exhalation, air in the caudoventral bronchi may jet dorsally (blue arrows in Fig. 1C) to enter the ostia of the dorsobronchi. In this way, a simple arrangement of the bronchi by themselves might give rise to unidirectional airflow. Also, the mechanism of gas exchange in crocodilians is not known; a crosscurrent mechanism has been hypothesized (11), but a countercurrent mechanism cannot be ruled out. Furthermore, the importance of unidirectional airflow for gas exchange efficiency in the alligator lung is not known and cannot be determined from our data, which consist of measurements of airflow.

Previous scenarios for the evolution of unidirectional airflow are that it arose in dinosaurs of coelurosaurian grade (12), convergently in theropods and pterosaurs (13, 15), or not at all in dinosaurs because of a hepatic piston mechanism of breathing (14). Our findings contrast with these previous views in several ways. They demonstrate that the hepatic piston mechanism of breathing, which crocodilians have but birds lack, does not preclude the evolution of unidirectional flow and that pneumaticity, which crocodilians lack, cannot be used to diagnose unidirectional airflow in fossil taxa, as previously suggested (13, 15). Crocodilians and birds are crown-group Archosauria. Therefore, in contrast to previous views, we suggest that unidirectional flow evolved before the divergence of crurotarsan and dinosaurian archosaurs and was present in the basal archosaurs and their descendants, including phytosaurs, aetosaurs, “rauisuchians,” and crocodylomorphs. The crurotarsans and, somewhat later, the dinosaurs supplanted the synapsids as the dominant members of the Triassic terrestrial vertebrate assemblage, with Triassic mammals existing as diminutive mouselike forms (16, 17). The roles of contingency and competition in the faunal turnover that occurred in the aftermath of the End Permian mass extinction are controversial. The basal archosaurs and archosauromorphs, animals such as Euparkaria, appear to have expanded their capacity for vigorous exercise (18) during a period of relative environmental hypoxia (19). In bird lungs, unidirectional airflow coupled with a crosscurrent mechanism of gas exchange facilitates the extraction of oxygen under conditions of hypoxia (20). If such a lung was present at the base of the archosaur radiation, this clade may have been better able than the synapsids to compete for niches that required a capacity for vigorous exercise.

Supporting Online Material

Materials and Methods

Fig. S1


Movies S1 to S3

References and Notes

  1. This work was supported by NSF (grant IOS-0818973 to C.G.F.).
View Abstract

Stay Connected to Science

Navigate This Article