Proposal (192) to South American Classification Committee


Split the genus Puffinus in two genera: Puffinus and Ardenna



The current SACC Checklist recognises:


Puffinus pacificus Wedge-tailed Shearwater 

Puffinus bulleri Buller's Shearwater 

Puffinus griseus Sooty Shearwater

Puffinus gravis Greater Shearwater 

Puffinus creatopus Pink-footed Shearwater 

Puffinus carneipes Flesh-footed Shearwater 

Puffinus puffinus Manx Shearwater 

Puffinus assimilis Little Shearwater 

Puffinus lherminieri Audubon's Shearwater 

Puffinus subalaris Galapagos Shearwater



Penhallurick and Wink carried out an analysis using Bayesian Inference (Mr Bayes3.1) of 95 taxa within the Procellariiformes.


We used PAUP (Swofford 1998) and ModelTest (Posada and Crandall 1998) to analyse the substitution patterns in a cytochrome-b data set against 56 different models of DNA evolution. We used the Akaike Information Criterion (AIC). The model selected was GTR+I+G.


We used Bayesian inference and Markov Chain Monte Carlo (mcmc) with Metropolis coupling for estimating phylogenetic hypotheses from DNA data. The model of nucleotide substitution used was the GTR model, following the results of ModelTest mentioned above. Thus Nst was set to 6 and Rates to Invgamma. In relation to prior distributions, we used a flat Dirichlet for both Revmatpr and Statefreqpr. Ronquist et al. (2005) stated that use of the default settings is appropriate if we want estimate these parameters from the data assuming no prior knowledge about their values. After three million generations, the average standard deviation of split frequencies was 0.008673. The mean of the 13 Potential Scale Reduction Factor (PSRF) figures was 1.002 (SD 0.003). We give, as Figure 1, a plot of the generation versus the log probability of the data (the log likelihood values). The plot approximates "white noise'. All the data just cited support the validity of the analysis.


Figure 1 Plot of generation versus the log probability of the data (the log likelihood values) for the MrBayes 3.1 analysis of the data


Figure 2 Phylogeny of all Procellariiformes

Bayesian Inference Tree. This represents a 50% majority rule consensus tree. Posterior Probability values are indicated, usually to the right of the nodes, but occasionally, for reasons of space, to the left.


Figure 3 displays that section of Figure 2 containing the shearwaters


Note: All of these documents can be found at the author's website

Click on Enter, and then on "SACC Documents" at the bottom of the menu on the left.


Figure 1 can be found under "Mr Bayes Plot"

Figure 2 under "Mr Bayes tree for Procellariiformes"

Figure 3 under "Mr Bayes subtree for Shearwaters"

and Table 1 under "Tamura-Nei Distance matrix for Shearwaters" 


It will be seen that the Shearwaters break into three distinct clades: a Calonectris clade, with a posterior probability of 1.00. The clade uniting Puffinus nativitatis Streets 1877 Christmas Shearwater with the remainder of the small Puffinus group has a posterior probability of 0.98. The clade comprising the larger shearwaters has a posterior probability of 1.00 in Figure 3.


There is clearly a division between the smaller and larger shearwaters. The question is: should this division be recognised at the generic or subgeneric level. Penhallurick and Wink (2004) argued that the distance data supported the recognition of distinct genera within each of these groups, although there has been a general bias against using distance data to make taxonomic judgments. Relevant here are the remarks of Helbig et al. (2002). On the role of nucleotide sequences in species judgments, Helbig et al. (2002: stated: "Molecular divergence is not a character (a particular sequence is), but sequence divergence measures can be used as an objective measure of overall divergence in comparative analyses. Molecular divergence, although not linearly correlated with phenotypic divergence, is proportional to the time that has elapsed since two taxa diverged from a common ancestor and thus gives a rough indication of how likely it is that reproductive incompatibilities have evolved between the two taxa.' (Emphasis added). We note the criteria outlined in this paper have been adopted by the Taxonomic Sub-Committee of the British Ornithologists' Union Records Committee (Sangster et al. 2004); and that the Taxonomic Advisory Committee in relation to the Agreement on the conservation of albatrosses and petrels had as the first of its proposed Plan of Action for the Taxonomy Working Group (Available at "Consider adopting the model presented by Helbig et al. (2002) ß to the taxa listed in the ACAP agreement'.


The within-groups Tamura-Nei distance for Puffinus (in the narrower sense) was 5.38% (SD 2.03); the within-groups mean for Ardenna was 4.75% (SD 1.51). The between-groups mean was 9.52%. (SD 0.08). Since the between-groups mean is nearly twice the within-groups mean, this suggests recognition of two distinct genera. Of further relevance is the distances between the three Calonectris species and the two groups: to the Ardenna group mean 10.51% (SD 0.56); to the smaller Puffinus group mean 9.62% (SD 0.80). Since the distances between Calonectris and the two groups approximates the distance between the two groups themselves, this is further evidence that the two groups should be differentiated at the generic level.


Wolters (1975-82) recognised two genera within the group of larger shearwaters (although griseus and tenuirostris remained in Puffinus): Ardenna Reichenbach 1853, to which he assigned gravis and carneipes, with creatopus as a subspecies; and Thyellodroma Stejneger 1888, to which he assigned pacificus and bulleri (1975-82: 36).


Recommendation: that the species currently within Puffinus in the South American Checklist be reassigned as follows:


Ardenna pacificus (J. F. Gmelin 1789) Wedge-tailed Shearwater 

Ardenna bulleri (Salvin 1888) Buller's Shearwater 

Ardenna griseus (J. F. Gmelin 1789) Sooty Shearwater

Ardenna gravis (O'Reilly 1818) Greater Shearwater 

Ardenna creatopus (Coues 1864) Pink-footed Shearwater 

Ardenna carneipes (Gould 1844) Flesh-footed Shearwater 

Puffinus puffinus (Brčnnich,1764) Manx Shearwater 

Puffinus assimilis (Gould 1838) Little Shearwater 

Puffinus lherminieri Lesson,1839 Audubon's Shearwater 

Puffinus subalaris Ridgway 1897 Galapagos Shearwater


I give notice that in a subsequent proposal, I will recommend the merging of Ardenna creatopus (Coues 1864) Pink-footed Shearwater Ardenna carneipes (Gould 1844) Flesh-footed Shearwater into a single species.



First meeting of the Advisory Committee in relation to the Agreement on the conservation of albatrosses and petrels (Available at Accessed 01/11/05

Helbig, A. J., Knox, A. G., Parkin, D. T., Sangster, G. and Collinson, M. (2002) Guidelines for assigning species rank. Ibis 144, 518-525.

del Hoyo, J., Elliott, A. and Sargatal, J. eds. (1992) 'Handbook of the Birds of the World', Vol. 1. (Lynx Edicions, Barcelona).

Penhallurick, J. M and Wink, M. (2004) Analysis of the taxonomy and nomenclature of the Procellariiformes based on complete nucleotide sequences of the mitochondrial cytochrome-b gene. Emu 104, 125-47.

Posada, D. and Crandall, K. A. (1998). MODELTEST: testing the model of DNA substitution. Bioinformatics 14, 817-88.

Ronquist, F., Huelsenbeck, J. P. and van der Mark, P. (2005) "MrBayes 3.1 Manual. Draft 26/05/2005'. Available at: Accessed 1/11/05

Sangster, G., Collinson, M., Helbig, A. J., Knox, A. G. and Parkin, D. T. (2004) Taxonomic Recommendations for British Birds: Second Report. Ibis146, 153-157.

Sibley, C. G. and Monroe, B. L. Jr. (1990) "Distribution and taxonomy of birds of the world' (Yale University Press, New Haven and London.)

Swofford, D. L. (1998) PAUP. Phylogenetic analysis using parsimony (and other methods). Version 4. Sinauer Associates. Sunderland, Massachusetts.


John Penhallurick, November 2005




Comments from Stiles: "Here, the question is less straightforward [than in 190 or 191]: whether to recognize Ardenna as a genus or subgenus. Helbig et al.'s statement relating probability of reproductive isolation to divergence time is not relevant to this question, since species status is not at issue. The question is, should genetic data alone (single gene) be considered sufficient for answer this question. My tentative feeling would be NO, but I would gladly accept illumination from other SACC members more familiar with such data and its possible limitations."


Comments from Remsen: "NO (barely). In this case (in contrast to 190 and 191) there is indeed a phenotypic character that is concordant with the phylogeny. However, I think body size per se is one of the least informative "characters" in birds -- I suspect we can all think of monophyletic genera that show a range of body size as great as that between Ardenna and Puffinus groups (Chloroceryle naturally pops into my mind). Nonetheless, I've gone back and forth on this one several times. John's restricted Puffinus group consists of all the notoriously similar "black"-and-white shearwaters but also includes (extralimital) dark-bellied nativitatis and heinrothi. With the current species rank of P. creatopus/P. carneipes based solely (I think) on underparts color, and their sister status regardless of rank unquestioned, color of underparts in shearwaters would seem to be hold little promise as a predictor of relatedness. Austin et al. (Auk 121; 854, 2004) also had genetic data that indicate that Ardenna and Puffinus (and Calonectris) are monophyletic groups, just as in Penhallurick and Wink, but those data are also cytochrome b, and only 917 bp.  Although I don't know these birds, I predict that Ardenna is worthy of recognition and that it and Puffinus-sensu stricto are natural groups, but I need just a little more data to push me to YES. Perhaps it already exists out there in the seabird literature?"


Comments solicited by Remsen from Dr. Carla Cicero, MVZ, UC Berkeley: "The only SACC person who has commented on the 3 proposals is Stiles. Gary is concerned (rightly so) about accepting these changes on the basis of a single mtDNA gene (cyt b). I agree, I think that is a problem with all three proposals, and if I were voting, I would probably vote "no" without additional congruent molecular evidence.


"I also have a problem with statements like the one in proposal 192, 'The within-groups Tamura-Nei distance for Puffinus (in the narrower sense) was 5.38% (SD 2.03); the within-groups mean for Ardenna was 4.75% (SD 1.51). The between-groups mean was 9.52%. (SD 0.08). Since the between-groups mean is nearly twice the within-groups mean, this suggests recognition of two distinct genera.' This is the barcoding mentality, where you reach a certain threshold of between vs. within taxon divergence and it suggests a new species, or new genus."


Comments solicited by Remsen from Dr. Carla Cicero, MVZ, UC Berkeley: "The cytochrome b phylogeny divides Puffinus into two clades that are different in body size. The consistent morphological differences lend additional credence to the cytochrome b topology in this case, and the body size differences combined with reciprocal monophyly seem sufficient to warrant generic recognition. I would like to see a slightly more detailed accounting of the body size distributions for each of the two Puffinus clades and any phenotypic synapomorphies that define each of them. For me, an mtDNA phylogeny can provide a strong basis for generic revision when it is concordant with phenotypic evidence... is that the case here?"


Comments from Robbins: "NO. To be consistent with my votes on proposal # 190 & 191 I vote NO."


Comments from Pacheco: "NO. Embora este caso seja um tanto diferente, prefiro aliar o meu voto com aquele das duas propostas precedentes na ausÉncia de evidÉncia molecular adicional."


Comments from Zimmer: "NO. Split genus Puffinus into two genera, Puffinus for smaller shearwaters, and Ardenna for larger ones. I think John Penhallurick is probably on the right track with this one, but as with earlier propositions, I am uncomfortable basing a taxonomic revision on genetic distances obtained from a single gene. It seems that phenotypic evidence corroborating the proposed phylogeny is out there, but hasn't been analyzed in a pertinent way. Until such time, I vote NO."


Comments from Jaramillo: "YES - Although I am being inconsistent here, as this data set is based on one gene as well, I am not troubled as I think it is healthy to be inconsistent once in a while. We are not machines, nor zealots here, but humans making a decision. This decision is less troubling to me than the last. There are biogeographical and biological issues that map to these two genera in an interesting way. The Ardenna, are all southern hemisphere breeders except for the tropical Wedge-tailed. The Puffinus include several exclusively Northern Hemisphere breeding taxa. I may be wrong but at breeding sites I do not think you ever find two species of Puffinus, or two species of Ardenna, but when you have two shearwaters breeding on the same island one is a Puffinus, and the other is an Ardenna. There may be some exceptions to this, but in general it works. One could argue that this is an issue of size and therefore competitive exclusion by the smaller and shorter winged "diving" Puffinus, and the larger and mostly surface feeding Ardenna. This is likely an important factor, but the development of this pattern is likely due to the original split into two clades and later colonization of the same breeding islands. I am perhaps speculating too much here, but my "gut" tells me that the genetic data match well to other aspects of these birds biology and distribution; therefore I am comfortable voting Yes on this one."