Proposal (412) to South American Classification Committee

Split Momotus momota into five species

This proposal would reverse the decision of Proposal no. 117 which favored lumping the highland form aequatorialis (considered a separate species in the baseline list) into a broad M. momota due to the lack of published evidence supporting the split, and in view of the fact that other taxa currently included in M. momota would probably deserve species rank were a comprehensive analysis to be performed. I have attempted such an analysis (Stiles 2009), now published in Ornitología Colombiana, the online journal of the Asociación Colombiana de Ornitología. Hopefully by now the members of SACC have received from Van a pdf of this study; if not, it can be downloaded from the web page of the journal Ornitología Colombiana: www.ornitologiacolombiana.org/revista/htm.

Basically, I drew my data from three sources: plumage patterns, biometrics and vocalizations, supplemented by information on geographic distributions and ecology. I examined a total of 512 specimens of ten “focal” taxa occurring in the area between southern Central America, northern and western South America from Colombia east to the Guianas and south to northern Peru, and Trinidad-Tobago. I also examined ca. 30 specimens from areas slightly to the south and east to further check for intraspecific variation. I defined 14 characters of plumage pattern and took six 6 measurements of bill, wing and tail. For vocalizations, I restricted the main analysis to the ‘hooting’ “primary song”; motmots have a much broader vocal repertoire but other vocalizations had not been recorded consistently for all taxa. From sonograms, I measured five parameters of frequency and duration for those taxa in which this song consisted of a single note, and six additional parameters for taxa in which the song consisted of two notes. Data were analyzed with t-tests, ANOVA, discriminant analysis and principal components analysis. I defined species limits in this complex on the basis of two general criteria: diagnosability and the probability that the differences observed would assure maintenance of reproductive isolation should currently allopatric groups enter into contact. My results support recognition of five species-level taxa in this complex: lessonii Lesson 1842 (including 2-3 additional subspecies in Mexico beyond the scope of this study), momota Linnaeus 1766 (including the nominate, microstephanus Sclater 1855 and several other subspecies of eastern and southern South America beyond the scope of this study); M. aequatorialis Gould 1857 (including the subspecies chlorolaemus Berlepsch and Stolzmann 1902); bahamensis Swainson 1837 and subrufescens Sclater 1853. In the latter species, I recognize as subspecies osgoodi Cory 1913, argenticinctus Sharpe 1892 and spatha Wetmore 1946, but find the following taxa not adequately diagnosable and recommend lumping them into nominate subrufescens: conexus Thayer & Bangs 1906, reconditus Nelson 1912 and olivaresi Hernandez & Romero 1978.

For the purposes of SACC, my analysis would recognize four species in our area (lessonii being restricted to Central America): cis-Andean momota, Andean aequatorialis, northwestern, trans-Andean subrufescens and Trinidad-Tobago bahamensis. My conclusions are congruent with a phylogeographic analysis of the Momotidae (as yet unpublished) by Chris Witt, save that bahamensis is nested within the subrufescens clade; I present arguments, mainly from plumage and biometrics, in support of species status for bahamensis. Regarding English names, I propose Amazonian Motmot for momota since the Amazon basin includes the vast majority of its distribution (and because of the great variation among the named subspecies, I could devise no adequately descriptive name suitable for all of them); Whooping Motmot for subrufescens because its rather long-drawn-out single-note song does indeed sound like a whoop; Andean Motmot for aequatorialis because it is indeed restricted to the Andes and because other species of motmot are also “highland” birds; and Trinidad Motmot for bahamensis.

The important references for this study are given in Proposal 117 and the pdf of this study. I recommend a YES on this proposal (obviously!).

F. Gary Stiles, August 2009

Comments from Robbins: “YES. Gary has thoroughly documented species level differences among these taxa.”

Comments from Zimmer: “YES. Gary has done a nice job of providing the analysis that we all wanted when we voted on Proposal 117. Biometrics, plumage patterns, and vocal data all point toward the proposed splits, and I would further add my support for Gary’s proposed English names for the various resulting species.”

Comments from Cadena: “YES. Gary has done an admirable job describing geographic variation in this group. Because many of the populations are allopatric, several difficulties remain regarding where does one draw species limits, but I think it is likely that these difficulties will persist regardless of how much additional data we throw at problems like this (a similar situation occurs in Arremon torquatus, on which I will submit a proposal shortly). Gary's proposed classification, which considers likelihood of reproductive isolation and also the distinctiveness of evolutionary lineages, is a substantial improvement in comparison to what we had before.”

Comments from Remsen: “YES. Gary has taken all available phenotypic data and partitioned the geographic variation into the units that are most defensible from the standpoint of known or likely reproductive isolation … a big step forward.”

Comments from Pacheco: “YES. im para a proposição em considerar momota, aequatorialis, subrufescens e bahamensis como espécies distintas. Gary fez um excelente trabalho elucidando as interrelações dos vários táxons de Momota presentes na região selecionada.”

Comments from Jaramillo: “YES. It is fantastic when a new classification is also a clarification. Traveling around it is clear that members of this group are certainly similar to each other, but at the same time the differences are notable. My first trip to Trinidad and Tobago had me staring at this strange thing, thinking…surely this is not the same creature as in Mexico, or Ecuador…or…. I particularly like that this is a new classification based on traditional methods, and it is tight and well done. It does scream out that while molecular methodology is an indispensable tool, you can attack these problems carefully with traditional datasets and come up with something very strong. I look forward to the eventual publication of molecular datasets on this, which will surely strengthen much of what is put forward here.”

Comments from Nores: “YES, pero con reservas. Aunque considero que el análisis hecho por Gary es excelente y tiene un detalle asombroso, hubiera sido perfecto 10 o 15 años atrás cuando no existían o estaban poco desarrollados los estudios meoleculares. En este momento, yo hubiera deseado ver algún análisis molecular antes de realizar la separación en cinco especies. Además, yo soy muy partidario del “biogeographic species concept” developed by Hellmayr: allopatric representatives of a common stock should be considered subspecies. A pesar de esto, considero que hasta tanto haya estudios moleculares está bien en aceptar la propuesta de separar las especies.”