SEARCH

SEARCH BY CITATION

Keywords:

  • acceleration;
  • Ardeidae;
  • deceleration;
  • endochondral ossification;
  • growth;
  • heterochrony;
  • heteroposy;
  • paedomorphosis;
  • peramorphosis

Abstract

  1. Top of page
  2. Abstract
  3. Introduction
  4. Material and methods
  5. Results
  6. Discussion
  7. Acknowledgments
  8. References

Evolutionary changes in developmental timing and rates (heterochrony) are a source of morphological variation. Here we explore a central issue in heterochronic analysis: are the alterations in developmental timing and rates the only factor underlying morphological heterochrony? Tarsometatarsal growth through endochondral ossification in Ardeidae evolution has been taken as a case study. Evolutionary changes in bone growth rate (morphological heterochrony) might be either (a) the result of alterations in the mitotic frequency of epiphyseal chondrocytes (process-heterochrony hypothesis), or (b) the outcome of alterations in the number of proliferating cells or in the size of hypertrophic chondrocytes (structural hypothesis). No correlation was found between tarsometatarsal growth rates and the frequency of cell division. However, bone growth rates were significantly correlated with the number of proliferating cells. These results support the structural hypothesis: morphological acceleration and deceleration are the outcome of evolutionary changes in one structural variable, the number of proliferating cells.


Introduction

  1. Top of page
  2. Abstract
  3. Introduction
  4. Material and methods
  5. Results
  6. Discussion
  7. Acknowledgments
  8. References

Evolutionary changes in developmental timing and rates (heterochrony) are responsible for most morphological variation within groups ( McKinney & McNamara, 1991; Reilly et al., 1997 ). The outcome of this evolutionary process is truncation (paedomorphosis) or extension (peramorphosis) of development giving a whole array of derived, but otherwise unrelated, character states ( Alberch & Blanco, 1996). The incidence of heterochrony in the evolution of a wide variety of organisms has been subject to detailed analysis after seminal papers by Gould (1977) and Alberch et al. (1979 ). However, to date, little attention has been paid to the developmental mechanisms underlying this evolutionary process ( Klingenberg, 1998). Here we address a central question in heterochronic analysis: are the alterations in developmental timing and rates the only factor underlying morphological heterochrony? To explore this issue, tarsometatarsal growth through endochondral ossification in Ardeidae evolution has been taken as a case study.

Ardeidae have been a subject of interest for evolutionary biologists in the last two decades from the point of view both of phylogenetic analysis ( Payne & Risley, 1976; Sheldon, 1987a, b; Sheldon & Kinnarney, 1993 ; McCracken & Sheldon, 1997, 1998) and of ecomorphology ( Boev, 1988, 1989). The highly diversified morphological patterns reported in Ardeidae ( Boev, 1988, 1989) strongly suggest that heterochronic processes were involved in different cladogenetic events during the evolution of the group. Here we analyse the incidence of pattern heterochronies in Ardeidae evolution, mainly following the method presented by Alberch et al. (1979 ) and modified by Reilly et al. (1997 ) and Cubo & Mañosa (1999). We also assess the underlying developmental mechanism of longitudinal bone growth, taking the scheme of Raff & Wray (1989) as a framework.

Endochondral ossification (a developmental process) accounts for longitudinal bone growth (the morphological outcome): physeal chondrocytes proliferate, they undergo hypertrophy and apoptosis and they are finally replaced by osteocytes transported by metaphyseal blood vessels ( Barreto et al., 1993 ; Karsenty, 1998). Bone longitudinal growth rate is a function of the number of proliferating cells, the frequency with which these cells divide and the size to which they grow prior to apoptosis ( Kirkwood & Kember, 1993). Hence, two hypotheses were considered: evolutionary changes in bone growth rate (morphological heterochrony) might be either the result of alterations in the frequency of cell division (process-heterochronyhypothesis), or the outcome of alterations in the number of proliferating cells or in the size of hypertrophic chondrocytes (structural hypothesis). Our results support the structural hypothesis: morphological acceleration and deceleration are the outcome of evolutionary changes in the number of proliferating cells.

Material and methods

  1. Top of page
  2. Abstract
  3. Introduction
  4. Material and methods
  5. Results
  6. Discussion
  7. Acknowledgments
  8. References

Morphological outcome: ontogenetic trajectories

For the study of ontogenetic trajectories we used 107 Ixobrychus minutus (Linnaeus 1766) (Little Bittern), 77 Ardeola ralloides (Scopoli 1769) (Squacco Heron), 13 Bubulcus ibis (Linnaeus 1758) (Cattle Egret), 24 Egretta garzetta (Linnaeus 1758) (Little Egret), 71 Ardea purpurea Linnaeus 1766 (Purple Heron) and 38 Ardea cinerea Linnaeus 1758 (Grey Heron). All growth data of nestlings were collected at the Ebro Delta (40°43′N, 0°42′E), with the only exception of those for A. cinerea, which were obtained at the Albufera of Valencia wetlands (37°40′N, 0°30′E). Chicks were weighed to the nearest gram using a spring balance (Pesola, Switzerland), and the length of their tarsometatarsus was measured to the nearest 0.01 mm using a digital caliper (Mitutoyo, Japan). Nestlings were marked on the first manipulation for subsequent recognition. An interval between measurements from 1 to 7 days was established for each species (one day for I. minutus and B. ibis, four for E. garzetta and A. ralloides, five for A. purpurea and seven for A. cinerea). Additionally, biometrical data from one chick of Nycticorax nycticorax (Linnaeus 1758) (Black-crowned Night Heron), measured at 3-day intervals, were further taken at the Medes Islands (42°3′N, 3°13′E) to be used in the histological study (see below). The visits were made at the same time of day to avoid errors caused by daily fluctuations in biological rhythms.

Interval growth curves were drawn for the body mass and for the tarsometatarsus length following the method described by Ricklefs & White (1975). This technique enabled us to obtain an average growth curve for a population (with relative and absolute chronological age scales) from growth increments of anatomical structures in a standardized time interval. Knowledge of the nestlings’ age is not necessary, but the sample must be heterogeneous enough to include nestlings at all the growth stages.

The average growth curve for each species was used to define the period of linear growth. Only data within this linear interval were then used to calculate the corresponding growth rate values at individual level. The distribution characteristics of this variable were analysed by means of Kolmogorov–Smirnov goodness of fit test (corrected by Lilliefor’s table) for normality and Levene tests for homogeneity of variances. Results indicated that a power transformation of data (xa; a = 0.021 for weight, a = 0.506 for tarsometatarsus) was required in order to improve the homoscedasticity of the samples ( Zar, 1984). The new derived matrices for weight and tarsometatarsus were checked with ANOVA and Tukey multiple comparison tests ( Zar, 1984; Sokal & Rohlf, 1995). Statistical analyses were performed using suitable software.

Developmental mechanism: cell kinetics of tarsometatarsus endochondral ossification

For histological studies we used the fresh corpses of chicks which we assumed to have died from natural causes (two N. nycticorax, one I. minutus, three A. ralloides, two B. ibis, two E. garzetta and four A. purpurea). The corpses were collected during the colony visits performed to obtain growth data at the Ebro Delta, with the only exception of N. nycticorax chicks, which were collected at the Medes Islands. For A. cinerea, given that no corpse was found, two live nestlings were collected under licence at the Albufera of Valencia wetlands. To avoid variation in the parameters of the cell kinetics of endochondral ossification linked to age, only animals that were in the linear period of growth were used in the histological analysis. These specimens were stored at –20 °C until the histological studies were carried out. Specimens were defrozen and dissected, and their tarsometatarsi were removed and fixed in formol saline. Tissue samples were taken from one tarsometatarsus of each specimen. Samples included complete sections of the proximal growth plate. They were decalcified in Zenker solution and embedded in paraffin wax. Longitudinal sections, 10 μm thick, were obtained. Sections were dewaxed in xylene, hydrated and stained with haematoxylin and eosin for analysis.

An exhaustive photographic record was obtained from three longitudinal sections of each growth plate in order to analyse the cell kinetics of endochondral ossification. In each cartilaginous growth plate there is a reserve cell zone, a proliferation zone, a zone in which cells undergo hypertrophy and maturation, and a zone of mature hypertrophied cells with calcified extracellular matrix ( Kirkwood & Kember, 1993). The longitudinal growth rate (Grl) of a long bone is a function of the following variables:

inline image

where n is the number of proliferating (flat) cells in both growth plates, f is the frequency with which these cells divide and h is the height of hypertrophic chondrocytes ( Kirkwood & Kember, 1993). Photomicrographs were obtained on a light microscope (Leitz Dialux, Germany) equipped with a camera (Wild MPS51, Switzerland). Two magnification levels were selected for different purposes. These were calculated empirically (×112 and ×285) from photographs of a stage micrometer which were taken at the beginning and at the end of each film as a calibration control. Images at the lowest magnification were used to measure the depth of the proliferation stratum of the proximal physis. Images at the highest magnification were used to measure the height of the proliferating and hypertrophic cells. Average cell heights were estimated by dividing transect length by cell count.

The number of proliferating cells in the proximal growth plate was obtained by dividing the depth of the proliferation strata by the mean heights of the flat cells. In order to obtain the total number of proliferating cells in both physes, data for the distal growth plate were estimated through the following equation: Distal cell number  =  –5.125 + Proximal cell number × 0.460, r2 = 0.880, n = 9 (the equation has been calculated with data taken from Kember et al., 1990 ). Finally, the mitotic frequency was obtained by using eqn 1: f = Grl/(n × h) ( Kirkwood & Kember, 1993).

Descriptive statistics were calculated for the height of proliferating chondrocytes, the number of these cells, the frequency with which these cells divide and the height of hypertrophic chondrocytes.

Developmental changes underlying morphological heterochrony

Correlations between bone growth rate and cell kinetics of endochondral ossification were carried out in order to assess the developmental basis of bone growth rate variation. The normality of the variables was evaluated through Kolmogorov–Smirnov and Shapiro–Wilks tests. The nonparametric Spearman correlation coefficient was used since normality tests were not significant in the case of the variation of mitotic frequency and the variation of height of hypertrophic chondrocytes.

Results

  1. Top of page
  2. Abstract
  3. Introduction
  4. Material and methods
  5. Results
  6. Discussion
  7. Acknowledgments
  8. References

Morphological outcome: ontogenetic trajectories

The average growth curves for the body mass and the tarsometatarsus length of all the species studied are shown in Figs 1 and 2. Descriptive statistics of the growth rates of the body mass and of the tarsometatarsus length from restricted samples (linear period of growth) for all the species studied are given in Table 1. Body mass growth rate ranged from 10.09 ± 4.21 g day–1 in I. minutus to 59.17 ± 17.46 g day–1 in A. cinerea. Tarsometatarsus growth rate ranged from 1.69 ± 0.68 mm day–1 in B. ibis to 4.23 ± 0.79 mm day–1 in A. cinerea. One-way analysis of variance for each biometrical variable was carried out for all the species studied. This analysis showed highly significant differences between species in all cases (P < 0.0001). Table 2 summarizes results from Tukey multiple comparison tests for all parameters. Most of the comparisons between species revealed significant differences (P < 0.05).

image

Figure 1.  Average growth curves for the body mass of the species analysed in this study. Abbreviations: Ac: Ardea cinerea; Ap: Ardea purpurea; Ar: Ardeola ralloides; Bi: Bubulcus ibis; Eg: Egretta garzetta and Im: Ixobrychus minutus.

Download figure to PowerPoint

image

Figure 2.  Average growth curves for the tarsometatarsus length of all the species studied. Abbreviations: Ac: Ardea cinerea; Ap: Ardea purpurea; Ar: Ardeola ralloides; Bi: Bubulcus ibis; Eg: Egretta garzetta and Im: Ixobrychus minutus.

Download figure to PowerPoint

Table 1.   Descriptive statistics of the body mass growth rates and tarsometatarsus longitudinal growth rates from restricted samples (linear period of growth) of all the species analysed in the morphological study. Thumbnail image of
Table 2.   Results of Tukey multiple comparison tests between the different species analysed in the morphological study for the body mass growth rate and the tarsometatarsus longitudinal growth rate. Significant differences (P < 0.05) are marked with an asterisk. N.S., not significant. Thumbnail image of

According to the phylogenetic analysis of McCracken & Sheldon (1998) based on osteological characters, the bitterns are the sister-group of a clade formed by day herons and tiger herons. I. minutus (a bittern) can be considered a sister-group of this clade. We assume that I. minutus retained the primitive condition for growth rate in body weight and tarsometatarsus length. On the other hand, within the clade day herons – tiger herons, A. ralloides is the sister-group of a clade that includes the larger species B. ibis, E. garzetta, A. purpurea and A. cinerea ( McCracken & Sheldon, 1998). Assuming this phylogenetic hypothesis, we analysed the occurrence of pattern heterochronies in Ardeidae evolution. For this, we tested whether the differences found in growth rates between species are isometric (i.e. only due to alterations linked to increases in body size, Cubo & Mañosa, 1999) or allometric (i.e. heterochrony due to a dissociation between the increases in age and size, Gould, 1977). To carry out this test, predictions on the isometric scaling of body mass growth rates and tarsometatarsus growth rates for the different species should be calculated from assumptions of the geometric similarity hypothesis ( Cubo & Casinos, 1997, 1998). According to this hypothesis, tarsometatarsus growth rates should be independent of adult body mass ( Kirkwood et al., 1989 ; Cubo & Mañosa, 1999), while body mass growth rates should scale to adult body mass raised to a power of 0.66 ( Cubo & Mañosa, 1999). We wondered whether these theoretical relationships hold for the sample analysed in this study. After testing for the normality of the variables through the Kolmogorov–Smirnov test, we found that: (1) tarsometatarsus growth rate was independent of adult body mass (the variables were not correlated: r = 0.768, P = 0.075), and the slope was not significantly different from the expected value of zero (P > 0.05) and (2) body mass growth rate was significantly correlated with adult body mass (r = 0.950, P = 0.004; Fig. 3) and the slope of the regression line was not significantly different from the expected value 0.66 (P > 0.05).

image

Figure 3.  Logarithmic plot of the regression of body mass growth rate to adult body mass of the species analysed (the latter taken from Cramp, 1977). Abbreviations: Ac: Ardea cinerea; Ap: Ardea purpurea; Ar: Ardeola ralloides; Bi: Bubulcus ibis; Eg: Egretta garzetta and Im: Ixobrychus minutus.

Download figure to PowerPoint

Isometric predictions on body mass growth rate in A. ralloides can be calculated through the general allometric equation of Huxley (1932):

inline image

where Gr is body mass growth rate, Mb is adult body mass and a and b are parameters. For this we need: (i) the expression on the isometric scaling of body mass growth rate discussed above ( Cubo & Mañosa, 1999): b = 0.66; and (ii) the plesiomorphic relative body mass growth rate (parameter a) calculated from I. minutus, a species that would have retained the plesiomorphic condition for the clade day herons – tiger herons:

inline image

where GrIm is the body mass growth rate and MbIm the adult body mass of I. minutus. So, the isometric body mass growth rate for A. ralloides (GrAr) could be estimated as follows:

inline image

where MbAr is the adult body mass of A. ralloides. On the other hand, assuming that tarsometatarsus growth rates are independent of adult body mass ( Kirkwood et al., 1989 ; Cubo & Mañosa, 1999), we predict that A. ralloides tarsometatarsus growth rate should not differ from the plesiomorphic condition (I. minutus, 3.11 mm day–1) (Prediction 2). Table 1 shows that A. ralloides body mass growth rate and tarsometatarsus longitudinal growth rate are significantly lower than predictions 1 and 2 (P < 0.05). These results are evidence for heterochrony in the evolution of A. ralloides relative to the plesiomorphic condition. According to the typology proposed by Reilly et al. (1997 ), we are facing a deceleration process, the morphological outcome of which is paedomorphosis.

A. ralloides branches basally in the clade day herons – tiger herons, and it is the sister-group of a clade that includes the larger species B. ibis,E. garzetta, A. purpurea and A. cinerea ( McCracken & Sheldon, 1998). Isometric predictions for B. ibis,E. garzetta, A. purpurea and A. cinerea body mass growth rates can be calculated assuming that the pattern of A. ralloides is the plesiomorphic condition. These predictions are as follows:

inline image

where GrBi, GrEg, GrAp and GrAc are the isometric body mass growth rate predictions for B. ibis,E. garzetta, A. purpurea and A. cinerea. Results from Table 1 show that empirical values of body mass growth rates are significantly higher than isometric predictions 3–6 (P < 0.05). These results suggest the occurrence of heterochrony, via acceleration, in the evolution of B. ibis,E. garzetta, A. purpurea and A. cinerea. These species share a derived global peramorphic morphology relative to the assumed plesiomorphic condition. On the other hand, under the isometric assumptions, tarsometatarsal growth rates of the derived species should not differ from that of the plesiomorphic condition (A. ralloides, 2.79 mm day–1, Prediction 7). Results (Table 1) show that while the tarsometatarsal growth of E. garzetta, A. purpurea and A. cinerea are accelerated and hence covaried with the global pattern of the species, B. ibis shows a growth mosaicism in which tarsometatarsal growth is decelerated and the overall growth is accelerated relative to the plesiomorphic condition.

Developmental mechanism: cell kinetics of tarsometatarsus endochondral ossification

Descriptive statistics of the number of proliferating chondrocytes, the height of these cells, the frequency with which these cells divide and the height of the hypertrophic chondrocytes of all the species analysed in the histological study are shown in Table 3. The number of proliferating cells shows a high degree of interspecific variation, ranging from 234 cells in I. minutus to 1027 ± 246 cells in A. cinerea. On the other hand, the frequency of mitosis ranges from 0.23 ± 0.06 divisions per day in B. ibis to 1.27 divisions per day in I. minutus. Finally, the interspecific variation of the height of the hypertrophic chondrocytes ranges from 10.54 ± 0.52 μm in A. ralloides to 12.18 ± 2.28 μm in E. garzetta.

Table 3.   Descriptive statistics of the cell kinetics of tarsometatarsus endochondral ossification of all the species analysed in the histological study. Thumbnail image of

Developmental changes underlying morphological heterochrony

To determine the developmental basis of bone growth rate variation (morphological outcome), correlations between this variable and the cell kinetics of endochondral ossification (developmental mechanism) were carried out. The Spearman correlation coefficient was not significant in the case of the mitotic rate and the tarsometatarsal growth rate (rs = 0.070, P = 0.797, n = 16). Nor was the size of hypertrophic cells correlated with tarsometatarsal growth rate (rs = 0.146, P = 0.589, n = 16). In contrast, the number of proliferating cells was significantly correlated with tarsometatarsus growth rate (rs = 0.548, P = 0.028, n = 16).

Discussion

  1. Top of page
  2. Abstract
  3. Introduction
  4. Material and methods
  5. Results
  6. Discussion
  7. Acknowledgments
  8. References

The selective process which limits the degree of similarity of closely related sympatric species has been evoked as an ultimate factor that accounts for interspecific body size variation ( Ruiz et al., 1998 ). In this study we present evidence for the occurrence of a number of heterochronies in Ardeidae evolution, the main morphological outcome of which has been the increase in body size of the derived peramorphic morphologies relative to the plesiomorphic pattern. According to the phylogenetic analysis of McCracken & Sheldon (1998) based on osteological characters, A. ralloides is the sister-group of a clade that includes the larger species B. ibis,E. garzetta, A. purpurea and A. cinerea. An evolutionary increase in overall size through acceleration has been identified in these species relative to A. ralloides, assuming that the latter has retained the plesiomorphic condition. On the other hand, in spite of the reported evidence for deceleration in the evolution of A. ralloides, this species is also larger than I. minutus, a species that would have retained the plesiomorphic condition. I. minutus is the sister-group of the clade day herons – tiger herons ( McCracken & Sheldon, 1998; osteological tree). The growth pattern of A. ralloides may be the product of a combination of heterochronic processes: (a) deceleration and (b) a prolongation of the growth period, which is 33% longer than in I. minutus ( Cramp, 1977). The occurrence of heterochronic phenomena in Ardeidae evolution could be ultimately caused by selection for decreasing the degree of similarity of closely related coexisting species (interspecific competition hypothesis, Diamond, 1978), but also by a relaxation of selection against a prolonged period of growth ( Lack, 1968). According to Lack (1968), the colonial breeding strategy of day herons would result in lower predation risk of their nestlings, thus enabling longer nestling periods relative to their noncolonial (I. minutus) sister-group.

The proximal factors that account for evolutionary morphological transformations can be studied at the many levels of biological organization ( Nelson, 1995). At the morphological (organ) level, it has been suggested that growth rate in birds is maximized and that it is limited by the functional maturity of tissues on the basis that there is a relationship between growth rate and mode of development within the altricial–precocial spectrum (see Ricklefs et al., 1998 , for a review). Therefore, within the same developmental mode, evolutionary increases in growth rate would be precluded ( Björklund, 1996). However, we report evidence for the occurrence of evolutionary acceleration in species of Ardeidae sharing the same semialtricial mode of development.

When the developmental proximal factors are considered, the frequency of cell division in tarsometatarsal growth plates of A. ralloides is lower than that of I. minutus (a species which would have retained the plesiomorphic condition, see above). In the case of B. ibis, the frequency of mitosis in tarsometatarsal growth plates is also lower than the plesiomorphic condition (which we assume to have been retained by A. ralloides, see above). In both cases, heterochronic changes in cellular mechanisms (evolutionary decreases in the mitotic frequency of tarsometatarsal chondrocytes) underlie morphological deceleration in tarsometatarsal growth rates (terminology from Reilly et al., 1997 ). On the other hand, the frequency of mitosis in tarsometatarsal growth plates of E. garzetta, A. purpurea and A. cinerea (derived morphologies, see above) are lower than that of A. ralloides (plesiomorphic condition). In these cases, heterochronic changes in cellular mechanisms are not congruent with morphological heterochrony, since evolutionary decreases in the mitotic frequency of tarsometatarsal chondrocytes underlie morphological acceleration of tarsometatarsal growth rates.

The paradox of developmental deceleration underlying morphological acceleration in the evolution of the day herons suggests the following question: what factors determine the interspecific variation of endochondral ossification parameters? Kember et al. (1990 ) found no evidence of a correlation between the frequency of mitosis of physeal chondrocytes and body mass in a diverse sample of birds. They concluded that the mitotic frequency is independent of the metabolic rate (which is dependent on body mass). In the case of the Ardeidae analysed in this study, the mitotic frequency of tarsometatarsal chondrocytes was not correlated with adult body mass (rs = – 0.331, P = 0.210, n = 16), and it appears to be independent of variation in metabolic rate. In relation with the scaling of the size of hypertrophic chondrocytes, the general hypothesis states that related animals of different size have cells of similar size ( Alexander, 1995). However, evidence is available for the alternative hypothesis that larger animals have larger cells ( Stevenson et al., 1995 ). In Ardeidae, the height of hypertrophic chondrocytes was not correlated with adult body mass (rs = 0.075, P = 0.784, n = 16). Therefore, these results are evidence for the general hypothesis that cell size is independent of body size. In fact, given that hypertrophic cell size varies within fairly narrow limits (see Table 3), we can consider it as a structural invariant in endochondral ossification. Finally, when the number of proliferating cells is considered, no scaling analysis can be carried out since it is a nondimensional variable. Alternatively, the thickness of the proliferating stratum of cells can be used, since, although it is the product of the number of proliferating cells and the height of these cells, cell height shows only a narrow range of variation (Table 3). The geometric similarity hypothesis predicts that linear measurements should scale to body mass raised to a power of 0.33 ( Cubo & Casinos, 1997, 1998); therefore, the thickness of the proliferating stratum of cells should scale likewise. Empirical results show that the depth of proliferating stratum of cells was significantly correlated with adult body mass (rs = 0.730, P = 0.001, n = 16), and the slope of the regression line was not significantly different from the expected value of 0.33 (P > 0.05). In conclusion, both the mitotic frequency and the height of hypertrophic chondrocytes are independent of adult body mass, and the thickness of the proliferating stratum of cells is the only developmental variable that scales with body size.

At this point, we address the central questions of this study: What are the relationships between proximal factors at different levels of biological organization? Are the alterations in developmental timing and rates the only factor underlying morphological heterochrony in Ardeidae evolution? Evolutionary changes in bone growth rate (morphological heterochrony) might be either the result of alterations in the frequency of cell division (process-heterochrony hypothesis) or the outcome of structural changes (alterations in the number of proliferating cells or in the size of hypertrophic chondrocytes, structural hypothesis). To test these hypotheses, correlations between tarsometatarsal growth rate and the cell kinetics of endochondral ossification were carried out. Tarsometatarsal growth rates were not correlated either with the mitotic rate of proliferating cells (P = 0.797) or with the height of hypertrophic chondrocytes (P =  0.589). In contrast, the number of proliferating cells was significantly correlated with tarsometatarsal growth rate (P = 0.028). Therefore, these results support the structural hypothesis: evolutionary changes in one structural variable, the number of proliferating cells at growth plates, underlie morphological acceleration/deceleration of tarsometatarsal growth rates. ‘Evolutionary changes in the abundance of a developmental process’ have been termed heteroposy ( Regier & Vlahos, 1988, pp. 20–21). Our results are therefore evidence for heteroposy in the evolution of the Ardeidae: we report evolutionary transformations in tarsometatarsal growth rates and morphology which are the outcome of changes in the amount of proliferating cells at growth plates. The concept of heteroposy should be added to Haeckel’s (1866) concepts of heterochrony (evolutionary changes of developmental timing and rates, Klingenberg, 1998) and heterotopy (evolutionary changes in the location of a developmental event, Wray & McClay, 1989), in order to complete our understanding of this kind of evolutionary phenomena.

The landmark papers of Alberch (1985), Hall (1984) and Raff & Wray (1989) pointed out the importance of the mechanistic developmental approach in heterochronic analysis. Raff & Wray (1989) presented examples of a nonheterochronic developmental basis for morphological heterochrony. In this study, we present a clear example of this phenomenon, and we conclude that developmental studies should be carried out in order to prevent the classification of very different evolutionary changes under the same category by using a phenomenological typology.

Acknowledgments

  1. Top of page
  2. Abstract
  3. Introduction
  4. Material and methods
  5. Results
  6. Discussion
  7. Acknowledgments
  8. References

Thanks are given to Dr M. Laurin and to R. Rycroft for improving the English text. We also thank J. Prosper for collaborating in the field work. We are grateful to the Spanish administration for giving us the scientific permits that allowed us to carry out the field work on the Ebro Delta, the Medes Islands and the Albufera of Valencia wetlands. J.C. has a postdoctoral grant from the ‘Ministerio de Educación y Cultura’ of Spain. The final development of this paper was funded by Generalitat de Catalunya Grant (1997SGR00158) to Consolidated Research Groups and grant PB95–0113-CO2–02 of the Spanish Government.

References

  1. Top of page
  2. Abstract
  3. Introduction
  4. Material and methods
  5. Results
  6. Discussion
  7. Acknowledgments
  8. References
  • 1
    Alberch, P. 1985. Problems with the interpretation of developmental sequences. Syst. Zool. 34: 46 58.
  • 2
    Alberch, P. & Blanco, M.J. 1996. Evolutionary patterns in ontogenetic transformation: from laws to regularities. Int. J. Dev. Biol. 40: 845 858.
  • 3
    Alberch, P., Gould, S.J., Oster, G.F., Wake, D.B. 1979. Size and shape in ontogeny and phylogeny. Paleobiology 5: 296 317.
  • 4
    Alexander, R. & McN. 1995. Big flies have bigger cells. Nature 375: 2020.
  • 5
    Barreto, C., Albrecht, R.M., Bjorling, D.E., Horner, J.R., Wilsman, N.J. 1993. Evidence of the growth plate and the growth of long bones in juvenile dinosaurs. Science 262: 2020 2023.
  • 6
    Björklund, M. 1996. Similarity of growth among Great tits (Parus major) and Blue tits (P. caeruleus) . Biol. J. Linn. Soc. 58: 343 355.
  • 7
    Boev, Z.N. 1988. Morphometrical characteristics of the adaptations of herons (Aves, Ardeidae) in relation to their movement along the substrate. Acta Zool. Bulg. 36: 63 71.
  • 8
    Boev, Z.N. 1989. Morphometrical characteristics of the adaptations in relation to the getting of food of the herons (Aves, Ardeidae). Acta Zool. Bulg. 37: 49 62.
  • 9
    Cramp, S. 1977. Handbook of the Birds of Europe, the Middle East and North Africa, the Birds of the Western Paleartic. Oxford University Press, Oxford.
  • 10
    Cubo, J. & Casinos, A. 1997. Flightlessness and long bone allometry in Palaeognathiformes and Sphenisciformes. Neth. J. Zool. 47: 209 226.
  • 11
    Cubo, J. & Casinos, A. 1998. Biomechanical significance of cross-sectional geometry of avian long bones. Eur. J. Morphol. 36: 19 28.
  • 12
    Cubo, J. & Mañosa, S. 1999. Evidence for heterochrony in the evolution of the goshawk Accipiter gentilis (Accipitridae, Aves). Ann. Sci. Nat. 20: 67 72.
  • 13
    Diamond, J.M. 1978. Niche shifts and the rediscovery of interspecific competition. Am. Sci. 66: 322 331.
  • 14
    Gould, S.J. 1977. Ontogeny and Phylogeny. The Belknap Press of Harvard University Press, Cambridge, Massachusset.
  • 15
    Haeckel, E. 1866. Generelle Morphologie der Organismen: Allgemeine Grundzüge der organischen Formen-Wissenschaft, mechanisch bergründet durch die von Charles Darwin reformirte Descendenz-Theorie. Reimer, Berlin.
  • 16
    Hall, B.K. 1984. Developmental processes underlying heterochrony as an evolutionary mechanism. Can. J. Zool. 62: 1 7.
  • 17
    Huxley, J.S. 1932. Problems of Relative Growth. Methuen, London. Reprinted 1993. Johns Hopkins University Press, Baltimore.
  • 18
    Karsenty, G. 1998. Genetics of skeletogenesis. Dev. Genet. 22: 301 313.DOI: 10.1002/(sici)1520-6408(1998)22:4<301::aid-dvg1>3.0.co;2-a
  • 19
    Kember, N.F., Kirkwood, J.K., Duignan, P.J., Godfrey, D., Spratt, D.J. 1990. Comparative cell kinetics of avian growth plates. Res. Vet. Sci. 49: 283 288.
  • 20
    Kirkwood, J.K., Duignan, P.J., Kember, N.F., Bennett, P.M., Price, D.J. 1989. The growth rate of the tarsometatarsus bone in birds. J. Zool., Lond. 217: 403 416.
  • 21
    Kirkwood, J.K. & Kember, N.F. 1993. Comparative quantitative histology of mammalian growth plates. J. Zool., Lond. 231: 543 562.
  • 22
    Klingenberg, C.P. 1998. Heterochrony and allometry: the analysis of evolutionary change in ontogeny. Biol. Rev. 73: 79 123.
  • 23
    Lack, D. 1968. Ecological Adaptations for Breeding in Birds. Methuen, London.
  • 24
    McCracken, K.G. & Sheldon, F.H. 1997. Avian vocalizations and phylogenetic signal. Proc. Natl. Acad. Sci. USA. 94: 3833 3836.
  • 25
    McCracken, K.G. & Sheldon, F.H. 1998. Molecular and osteological heron phylogenies: sources of incongruence. Auk 115: 127 141.
  • 26
    McKinney, M.L. & McNamara, K.J. 1991. Heterochrony: the Evolution of Ontogeny. Plenum, New York.
  • 27
    Nelson, R.J. 1995. An Introduction to Behavioral Endocrinology. Sinauer, Sunderland, Massachusetts.
  • 28
    Payne, R.B. & Risley, C.J. 1976 Systematics and evolutionary relationships among the herons (Ardeidae). Misc. Publ. Univ. Michigan Mus. Zool. 150: 1 115.
  • 29
    Raff, R.A. & Wray, G.A. 1989. Heterochrony: developmental mechanisms and evolutionary results. J. Evol. Biol. 2: 409 434.
  • 30
    Regier, J.C. & Vlahos, N.S. 1988. Heterochrony and the introduction of novel modes of morphogenesis during the evolution of moth choriogenesis. J. Mol. Evol. 28: 19 31.
  • 31
    Reilly, S.M., Wiley, E.O., Meinhardt, D.J. 1997. An integrative approach to heterochrony: the distinction between interspecific and intraspecific phenomena. Biol. J. Linn. Soc. 60: 119 143.
  • 32
    Ricklefs, R.E., Starck, J.M., Konarzewski, M. 1998. Internal constraints on growth in birds. In: Avian Growth and Development. Evolution Within the Altricial-Precocial Spectrum (J. M. Starck & R. E. Ricklefs, eds), pp. 266–287. Oxford University Press, New York.
  • 33
    Ricklefs, R.E. & White, S.C. 1975. A method for constructing nestling growth curves from brief visits to seabird colonies. Bird Banding 46: 135 140.
  • 34
    Ruiz, X., González-Solís, J., Oro, D., Jover, L. 1998. Body size variation in Audouin’s Gull Larus adouinii; a density-dependent effect? Ibis 140: 431 438.
  • 35
    Sheldon, F.H. 1987a. Phylogeny of herons estimated from DNA-DNA hybridization data. Auk 104: 97 108.
  • 36
    Sheldon, F.H. 1987b. Rates of single-copy DNA evolution in herons. Mol. Biol. Evol. 4: 56 69.
  • 37
    Sheldon, F.H. & Kinnarney, M. 19930. The effect of sequence removal on DNA-hybridization estimates of distance, phylogeny, and rates of evolution. Syst. Biol. 42: 32 48.
  • 38
    Sokal, R.R. & Rohlf, F.J. 1995. Biometry: the Principles and Practice of Statistics in Biological Research, 3rd edn. W. H. Freemann, New York.
  • 39
    Stevenson, R.D., Hill, M.F., Bryant, P.J. 1995. Organ and cell allometry in Hawaiian Drosophila: how to make a big fly . Proc. R. Soc. Lond. B. 259: 105 110.
  • 40
    Wray, G.A. & McClay, D.R. 1989. Molecular heterochronies and heterotopies in early echinoid development. Evolution 43: 803 813.
  • 41
    Zar, J.H. 1984. Biostatistical Analysis. Prentice Hall International, London.