Evaluating phenotypic plasticity in attachment organs of parasites can provide information on the capacity to colonize new hosts and illuminate evolutionary processes driving host specificity.We analysed the variability in shape and size of the dorsal and ventral anchors of Ligophorus cephali from Mugil cephalus by means of geometric morphometrics and multivariate statistics. We also assessed the morphological integration between anchors and between the roots and points in order to gain insight into their functional morphology. Dorsal and ventral anchors showed a similar gradient of overall shape variation, but the amount of localized changes was much higher in the former. Statistical models describing variations in shape and size revealed clear differences between anchors. The dorsal anchor/bar complex seems more mobile than the ventral one in Ligophorus, and these differences may reflect different functional roles in attachment to the gills. The lower residual variation associated with the ventral anchor models suggests a tighter control of their shape and size, perhaps because these anchors seem to be responsible for firmer attachment and their size and shape would allow more effective responses to characteristics of the microenvironment within the individual host. Despite these putative functional differences, the high level of morphological integration indicates a concerted action between anchors. In addition, we found a slight, although significant, morphological integration between roots and points in both anchors, which suggests that a large fraction of the observed phenotypic variation does not compromise the functional role of anchors as levers. Given the low level of genetic variation in our sample, it is likely that much of the morphological variation reflects host-driven plastic responses. This supports the hypothesis of monogenean specificity through host-switching and rapid speciation. The present study demonstrates the potential of geometric morphometrics to provide new and previously unexplored insights into the functional morphology of attachment and evolutionary processes of host-parasite coevolution.4

The search for phylogenetic signal in morphological traits using geometric morphometrics represents a powerful approach to estimate the relative weights of convergence and shared evolutionary history in shaping organismal form. We assessed phylogenetic signal in the form of ventral and dorsal haptoral anchors of 14 species of Ligophorus occurring on grey mullets (Osteichthyes: Mugilidae) from the Mediterranean, the Black Sea and the Sea of Azov. The phylogenetic relationships among these species were mapped onto the morphospaces of shape and size of dorsal and ventral anchors and two different tests were applied to establish whether the spatial positions in the morphospace were dictated by chance. Overall significant phylogenetic signal was found in the data. Allometric effects on anchor shape were moderate or non-significant in the case of evolutionary allometry. Relatively phylogenetically distant species occurring on the same host differed markedly in anchor morphology indicating little influence of host species on anchor form. Our results suggest that common descent and shared evolutionary history play a major role in determining the shape and, to a lesser degree in the size of haptoral anchors in Ligophorus spp. The present approach allowed tracing paths of morphological evolution in anchor shape. Species with narrow anchors and long shafts were associated predominately with Liza saliens. This morphology was considered to be ancestral relative to anchors of species occurring on Liza haematocheila and M. cephalus possessing shorter shafts and longer roots. Evidence for phylogenetic signal was more compelling for the ventral anchors, than for the dorsal ones, which could reflect different functional roles in attachment to the gills. Although phylogeny and homoplasy may act differently in other monogeneans, the present study delivers a common framework to address effectively the relationships among morphology, phylogeny and other traits, such as host specificity or niche occupancy.

An important question in the study of phenotypic evolution is whether characters are independent of each other or behave and evolve as integrated modules. Morphological integration and modularity provide a powerful framework for the analysis of the evolution of morphological traits. We used geometric morphometrics and phylogenetically independent contrasts (PIC) to test four different modularity hypotheses in the haptoral anchors of 14 monogenean species of Ligophorus. Integration between the modular units identified was further evaluated with two-block partial least squares analysis. Roots and points represented two modules in the dorsal and ventral anchors, but modularity was not statistically supported when parasite phylogeny was accounted for, which may indicate convergent evolution related to host characteristics and gill morphology. In contrast, PIC revealed medial and lateral modules in ventral anchors only. Moreover, we found evidence for ventral and dorsal anchor pairs forming two modules, supporting the notion that they play different functional roles. Integration between all identified modules was strong. We conclude that there is modular structure in the anchors of Ligophorus spp., accounted by adaptive and phylogenetic factors acting at different levels, and ventral and dorsal anchors evolve as integrated modules with specific roles in attachment.

Phenotypic variation results from the balance between sources of variation and counteracting regulatory mechanisms. Canalization and developmental stability are two such mechanisms, acting at two different levels of regulation. The issue of whether or not they act concurrently as a common developmental buffering capacity has been subject to debate. We used geometric morphometrics to quantify the mechanisms that guarantee phenotypic constancy in the haptoral anchors of Ligophorus cephali. Canalization and developmental stability were appraised by estimating inter- and intra-individual variation, respectively, in size and shape of dorsal and ventral anchors. The latter variation was estimated as fluctuating asymmetry (FA) between anchor pairs. The general-buffering-capacity hypothesis was tested by two different methods based on correlations and Principal Components Analyses of the different components of size and shape variation. Evidence for FA in the dorsal and ventral anchors in both shape and size was found. Our analyses supported the hypothesis of a general developmental buffering capacity. The evidence was more compelling for shape than for size and, particularly, for the ventral anchors than for the dorsal ones. These results are in line with previous studies of dactylogyrids suggesting that ventral anchors secure a firmer, more permanent attachment, whereas dorsal anchors are more mobile. Because fixation to the host is crucial for survival in ectoparasites, we suggest that homeostatic development of the ventral anchors has been promoted to ensure the morphological constancy required for efficient attachment. Geometric morphometrics can be readily applied to other host-monogenean models, affording not only to disentangle the effects of canalization and developmental stability, as shown herein, but to further partition the environmental and genetic components of the former.

Resumen. Se caracterizaron las comunidades componentes e infracomunidades de helmintos de Symphurus plagiusa en 37 puntos de muestreo de mayo a septiembre de 2003 en la costa de Campeche. Además, se exploraron posibles asociaciones estadísticas entre la abundancia de cada especie y los parámetros ambientales (salinidad, oxígeno, volumen de descarga de agua de ríos) mediante análisis de redundancia (RDA). Se examinaron 531 hospederos, cuyas comunidades componentes e infracomunidades fueron pobres en número de especies (máximos: 8 y 3 ± 1 respectivamente), pero no en individuos (16775 y 47 ± 32). La diversidad fue baja (componente: Simpson = 1; infracomunidad: Brillouin = 0.24 ± 0.11). La similitud en cada mes fue baja para componente (cualitativa, Jaccard: 0.25 ± 0.34; cuantitativa, % de similitud: 51 ± 49) e infracomunidad (cualitativa: 0.59 ± 0.24; cuantitativa: 46 ± 24). La similitud entre meses fue alta en componente (cualitativa: 0.63 ± 0.28; cuantitativa: 82 ± 15) e infracomunidad (cualitativa: 0.89 ± 0.10; cuantitativa: 77 ± 21). Hubo asociaciones signifi cativas entre la abundancia de especies y el volumen de descarga de agua de los ríos entre meses, lo que sugiere que este parámetro infl uye fuertemente sobre la composición y estructura de estas comunidades. Palabras clave: comunidades de helmintos, variación temporal, lenguado, golfo de México.