Multifunctionality is often framed as a core constraint of phenotypic evolution. Mouthbrooding, a form of parental care where offspring develop inside a parent’s mouth, increases multifunctionality by adding a major function (reproduction) to a structure already serving other vital functions (feeding and respiration). Despite increasing multifunctionality, mouthbrooding has evolved repeatedly from other forms of parental care in at least 7 fish families. We hypothesized that mouthbrooding is more likely to evolve in lineages with feeding adaptations that are already advantageous for mouthbrooding. We tested this hypothesis in Neotropical cichlids, where mouthbrooding has evolved 4–5 times, largely within winnowing clades, providing several pairwise comparisons between substrate brooding and mouthbrooding sister taxa. We found that the mouthbrooding transition rate was 15 times higher in winnowing than in non-winnowing clades, and that mouthbrooders and winnowers overlapped substantially in their buccal cavity morphologies, which is where offspring are incubated. Species that exhibit one or both of these behaviors had larger, more curved buccal cavities, while species that exhibit neither behavior had narrow, cylindrical buccal cavities. Given the results we present here, we propose a new model for the evolution of mouthbrooding, integrating the roles of multifunctional morphology and the environment.

Phylogeny: We used a 559-species Neotropical cichlid phylogeny trimmed from the cichlid phylogeny presented in McGee et al. (2020).

Behavioral classifications: We classified every species in this tree for which data was available as either a winnower or a non-winnower, and as either a mouthbrooder or a substrate brooder, so that every species was categorized as one of four possible states: 1) mouthbrooding and winnowing; 2) mouthbrooding and non-winnowing; 3) substrate brooding and winnowing; or 4) substrate brooding and non-winnowing. Parental care data were gathered from Stawikowski & Werner (1998), Goodwin et al. (1998), and López-Fernández et al. (2012), and winnowing data from López-Fernández et al. (2014), Weller et al. (2017), and López-Fernández et al. (2012). Because winnowing and mouthbrooding encompass a range of behaviors, we classified species inclusively for each category: a species was classified in a behavioral category if individuals exhibit the behavior to any degree that it could impact their morphology. The variations of each behavior (for example, mouthbrooding eggs or mouthbrooding larvae) probably pose slightly different constraints, but because we lacked sufficient phylogenetic and behavioral data to differentiate these fine-scaled differences, here we focus on their aggregate effects. Any species in which at least one sex orally incubates the offspring during at least one developmental period (eggs, larvae, or juveniles) was classified as a mouthbrooder. Similarly, any species which feeds by orally sifting substrate for invertebrates or edible detritus was classified as a winnower, even if it also feeds using other strategies. If we could not reliably determine whether or not a species is a mouthbrooder--for example, if it was most commonly reported as substrate brooding and only rarely reported as mouthbrooding in aquaria (Staeck 2015), or if mouthbrooding and substrate brooding behaviors seemed highly variable even in captivity (Römer et al. 2017, Breeze 2007), these species were conservatively classified as substrate brooders (4 species total). We also ran our comparative analyses below on versions of the dataset with 1) these species classified as mouthbrooders, and 2) excluded from the dataset altogether, to assess the sensitivity of our results to these ambiguous cases.

Dissections: Dissection images include 41 species across 20 genera representing the major radiations of Neotropical cichlids, mostly from the University of Michigan Museum of Zoology and the Field Museum of Natural History. Minimally, we sampled to the resolution of the node representing the most recent common ancestor between mouthbrooding and non-mouthbrooding taxa for each clade of mouthbrooding species, so that we sampled the immediate substrate brooding sister taxon for each mouthbrooding taxon. We selected only specimens that were fixed in closed-mouth, neutral positions (i.e. no opercular flaring or hyoid depression) for analysis. Each specimen was dissected using a modified version of the procedure described by Ridewood (1904) to remove the right suspensorium and oral jaw elements, exposing the buccal cavity and branchial arches. First, we made a midline cut on an anterior-posterior axis separating the premaxillary bones, and continued this cut by bisecting the parasphenoid along the midline. We separated the circumorbital bones from around the orbit before cutting the levator operculi. Ventrally, we separated the basihyal from its articulation with the urohyal and basibranchials, then cut the branchiostegal membrane from the urohyal through the mandibular symphysis. We were then able to remove the entire right cheek intact, including the opercular series, suspensorium, adductor mandibulae, nasal and circumorbital bones, and oral jaws.

We photographed specimens before and after dissection on either a Leica or Zeiss stereomicroscope to capture both external and buccal cranial morphologies. We based our external landmarks on those in Weller (2017), including curves for the rostrum, eye, lower jaw, operculum, and adductor mandibulae complex. For the buccal cavity, we chose landmarks describing structures that make up the borders of the cavity, including the vomer, parasphenoid, basibranchial series, hyoid, and lower jaw. A full list of fixed and sliding landmarks is given in the supplemental information.

Landmarks: Photographs were scaled and landmarked using StereoMorph (Olsen 2015) then imported into geomorph using the readland.shapes function, which converts StereoMorph curves into sliding semilandmarks (Adams et al. 2013).

Consult the README.md file for the specific contents of each code and datafile and how to run them.