SEARCH

SEARCH BY CITATION

Keywords:

  • age determination;
  • small odontocete species;
  • histological techniques

Abstract

  1. Top of page
  2. Introduction
  3. Materials and Methods
  4. Results
  5. Discussion
  6. Acknowledgments
  7. Literature Cited

Age estimation in odontocetes is based on counts of growth layer groups (GLGs) deposited in recording structures such as teeth. Generally, tooth sections are obtained using a cryostat microtome. However, some researchers prefer obtaining thin sections using a traditional paraffin microtome. Little information is available on the application of this technique to dolphin teeth. Our main aim was to investigate if the paraffin technique can be a viable alternative. We considered whether estimated age would be affected by preparation technique, staining method, and section thickness, while controlling for effects of species, body length, and sex. We also analyzed whether the staining method would affect readability of GLGs and age reading variability. Teeth from 86 individuals (representing seven species) were used, but not all were prepared using both techniques because sufficient teeth were not available in all cases. Although the staining method had significant effects on the estimated age using both techniques, the variability of GLG counts was small and appeared to be similar for both techniques. Using Mayer's hematoxylin stained sections of 8 μm thickness, good agreement of ages was obtained from both techniques, with more preparations classified as “good quality” for the paraffin technique. Mayer's hematoxylin provided the best contrast of the GLGs when using the paraffin technique. We conclude that the paraffin technique is viable and represents a cost-effective alternative to a cryostat microtome when preparing cetacean teeth for age determination.


Introduction

  1. Top of page
  2. Introduction
  3. Materials and Methods
  4. Results
  5. Discussion
  6. Acknowledgments
  7. Literature Cited

Age determination is a fundamental prerequisite for interpreting many aspects of the biology, ecology, and physiology of marine mammals. The dynamics of a population cannot be determined without accurate information on age composition, age at sexual maturity, age at first reproduction, and natural longevity (Myrick et al. 1983, Hohn 2002). Furthermore, knowledge of age composition provides essential information for estimating fecundity or mortality rates (Hohn 2002).

Age estimation in odontocetes is based on counts of growth layers groups (GLGs) deposited in recording structures such as teeth (dentine and cementum) and bone tissue (Perrin and Myrick 1980, Hohn 2002). Teeth of most odontocetes are homodont and monophyodont (i.e., one set of permanent teeth that are all similar) with growth layers being deposited continuously throughout life (Lockyer 1995, Hohn 2002). Dentine constitutes most of the tooth and is the tissue most often used for estimating age in odontocetes (Perrin and Myrick 1980, Klevezal 1996, Hohn 2002). Cementum covers the tooth root and is also used to estimate ages of older animals in which dentinal age estimates are not possible (Klevezal 1996, Hohn 2002). A GLG is “a group of layers deposited parallel to the formative surface of a tissue that occur with cyclical and predictable repetition” (Perrin and Myrick 1980) and generally consists of a broad opaque layer and an adjacent narrow translucent layer when viewed under transmitted light, or alternating stainable and unstainable layers in decalcified and stained sections (Perrin and Myrick 1980, Hohn 2002). The observed contrast is due to differences in content and distribution of the mineral component, resulting in differences in optical density and stainability, with the more transparent bands, except for the neonatal line, containing more mineral compounds (Klevezal 1996, Hohn 2002). The layers are thought to be formed as a result of seasonal changes in the growth rate of the tooth, which is correlated to seasonal changes in the growth rate of the animal (Klevezal 1980). Thus, the broad layer is thought to record a period of rapid growth in spring–summer and the narrow layer forms in autumn–winter when growth is slower (Klevezal 1980, Myrick and Cornell 1990).

The annual deposition of GLGs has been verified in several cetacean species, including both captive and wild bottlenose dolphins (Tursiops truncatus) (Hohn 1980b, Hohn et al. 1989), short-finned pilot whales (Globicephala macrorhynchus) (Lockyer 1993), spinner dolphins (Stenella longirostris), common dolphins (Delphinus delphis) (Collet 1981), and harbor porpoises (Phocoena phocoena) (Nielsen 1972), based on calibration studies (e.g., using tetracycline) (Gurevich et al. 1980, Myrick et al. 1984, Hohn et al. 1989, Myrick and Cornell 1990). Within taxonomic groups, (e.g., the delphinids), the layering patterns of different species show many similarities (Hohn 1990, Lockyer 1995, Hohn 2002, Hohn and Pinedo 2000) and it is thought that calibrations obtained for one species can be applied to related species. Nevertheless, GLG is a generic term and its equivalence to an annual growth increment needs to be determined in each instance of use (Perrin and Myrick 1980).

Counting of GLGs in teeth has been the most widely adopted method for age determination in odontocetes. Over the years, many preparation techniques have been developed (see Perrin and Myrick 1980), some of which have proved unsatisfactory in revealing GLGs (Perrin et al. 1977), or provide excellent resolution of GLGs, but are too time consuming or expensive to be applied to large samples of teeth (Hohn 1980b). Hohn and Fernández (1999) suggested that the tooth preparation technique and the method used to section teeth could also introduce biases into the interpretation of age. The layering pattern in a tooth can be complex due to the presence of additional elements inside a layer such as the presence of accessory layers, making it more difficult to identify a GLG (Hohn 1980a, Klevezal 1996).

Most researchers working with decalcified teeth have obtained sections at 15–20 μm thicknesses by using a freezing microtome (e.g.,Perrin and Myrick 1980, Hohn and Lockyer 1995). This method is preferable when it is necessary to cut pieces of hard tissues >1 cm thick. It is also possible to obtain thin sections of decalcified teeth using a cryostat, which is a combination of a refrigerator with a microtome. This technique is more time-consuming than using a freezing microtome, although it is extensively used for preparing dolphin teeth (Klevezal 1996). However, others prefer standard paraffin methods and prepare sections following standard microtechniques used with soft tissues (e.g., Humason 1962). In this way thin sections of decalcified teeth are obtained using a “normal” or paraffin microtome. Most studies in which this technique has been applied refer to terrestrial mammals and very little information has been published on the application of this technique in dolphin teeth. For example, Slooten (1991), working on teeth of Hector's dolphin (Cephalorhynchus hectori), suggested that it was possible to simplify the traditional procedure for preparing dolphin teeth by adopting standard procedures used on soft tissues. In her study, once decalcification was complete, about one-third was cut off each tooth longitudinally to hasten the process of reaching the pulp cavity during sectioning. Teeth were placed in separate plastic cassettes and put through a standard tissue processing and paraffin embedding process, as used for soft tissues (e.g., Humason 1962, Slooten 1991). Then dolphin teeth were sectioned at 2–4 μm using a standard microtome. Although the technique satisfactorily revealed the GLGs, no comparative analysis among different techniques using the same material was carried out to explore whether the resulting estimated ages might differ between techniques and which technique might be generally most reliable.

In a more recent study, Duignan and Jones (2005) prepared Hector's dolphin teeth according to Slooten's (1991), protocol although they tried to obtain thicker tooth sections (at 10–20 μm). Some technical problems were encountered with the sectioning equipment and, for several animals, tooth sections were not adequate to determine the age with accuracy. After wax embedding, some teeth became harder to section using the standard microtome.1 To date, Hector's dolphin is the only species in which the paraffin technique has been applied with limited success for preparing teeth and there is no published work about the use of wax embedding in other small cetacean species such as harbor porpoises or common dolphins.

The main aim of the present study was to investigate whether the paraffin technique can be applied as a viable alternative to the cryostat technique for preparing dolphin tooth sections for age determination. Furthermore, within each technique, we compared different staining methods and/or section thicknesses in order to see whether these factors contributed significantly to the quality (readability) of sections in terms of the contrast of the GLGs, the variability associated with age readings and the estimated age. We also tested for differences between species (as well as effects of body length and sex) by using teeth from several species of small cetaceans to investigate these questions. No teeth from known-age animals were available so, when possible, we applied the various methods to duplicate teeth from the same set of animals to facilitate comparisons.

Materials and Methods

  1. Top of page
  2. Introduction
  3. Materials and Methods
  4. Results
  5. Discussion
  6. Acknowledgments
  7. Literature Cited

Samples for Age Determination

Teeth were acquired from small cetaceans belonging to nine species (see Table 1 for details) stranded in two geographical areas of the northern Atlantic, Scotland and the Canary Islands. The material was collected by the local stranding response groups, coordinated by the Scottish Agricultural College (SAC) Veterinary Services in Inverness (Scotland) and by the Institute for Animal Health at the University of Las Palmas Gran Canaria (Spain), respectively. During the postmortem examinations, at least 4–5 teeth were removed from the middle of the lower jaw (Kuiken and Hartmann 1991) and fixed in 10% neutral buffered formalin (for about 24 h) before processing.

Table 1.  Summary of total number of teeth processed, by species and geographical area (M = male, F = female).
SpeciesCodeScotlandCanary IslandsTotal
Harbor porpoise (Phocoena phocoena)Pp20 M, 16 F36
Common dolphin (Delphinus delphis)Dd3 M4 F 7
Atlantic white-sided dolphin (Lagenorhynchus acutus)Lac3 M, 3 F 6
Striped dolphin (Stenella coeruleoalba)Sc1 M8 M, 1 F10
White-beaked dolphin (Lagenorhynchus albirostris)Lal5 M 5
Atlantic spotted dolphin (Stenella frontalis)Sf3 M, 6 F 9
Bottlenose dolphin (Tursiops truncatus)Tt5 M, 5 F10
Pygmy sperm whale (Kogia breviceps)Kb2 M 2
Dwarf sperm whale (Kogia sima)Ks1 M 1
Total 513586

For this study, teeth from 86 animals were used although not all teeth were processed using both techniques since sufficient teeth were not available in all cases. Therefore, the material was selected according to availability (Table 1 for details). For smaller species, a minimum of two teeth was required per individual (one per technique) since both techniques could not be applied to the same tooth. In harbor porpoises, because teeth were sectioned at two different orientations (i.e.,“porpoise” and “dolphin” cuts, respectively), four teeth were needed per individual (two teeth per technique). For those species with large teeth such as bottlenose dolphin and pygmy sperm whale (Kogia breviceps), the same tooth was prepared using both techniques. Thus, the tooth was first cut along the mid-longitudinal plane obtaining two halves and then each half was prepared by one of the two techniques. An additional tooth was required when it was not possible to obtain good tooth sections.

Tooth-Preparation Techniques

Decalcification— After 24 h in 10% formalin, teeth were gently rinsed in running water and placed in perforated plastic baskets for decalcification. Teeth were decalcified using RDO, a commercial rapid decalcifying agent (hydrochloric acid is the principal active ingredient) until they were flexible enough to section. Using RDO, decalcification time ranged from 4 to 20 h. Once teeth were rubbery in texture, they were rinsed in running tap water for several hours. The length of time required for decalcification varied between samples in relation to tooth size, age, and species, with longer times required for larger and older individuals. Teeth from small delphinids, including those from Atlantic spotted dolphin (Stenella frontalis), striped dolphin (Stenella coerulealba), common dolphin, harbor porpoise, Atlantic white-sided dolphin (Lagenorhynchus acutus), and white-beaked dolphin (Lagenorhynchus albirostris) were decalcified whole, whereas larger teeth from bottlenose dolphin and pygmy sperm whale took longer to be decalcified. The latter teeth were first cut along the mid-longitudinal plane using a Buehler Isomet low-speed diamond saw, obtaining wedges from the tooth that were decalcified separately.

Sectioning— The sectioning equipment differed between the two techniques. The cryostat technique was adapted from Hohn and Lockyer (1995). Teeth from all 86 individuals were prepared using this technique and after being rinsed in running tap water, they were mounted on freezing blocks with O.C.T (Sakura Finetek) compound and sectioned longitudinally using a freezing (−10°C) microtome (TE Electronic Cryostat). For 67 of these 86 individuals, sufficient teeth were available to prepare sections of 8 μm thickness, while for only 30 of these 86 individuals teeth were sectioned at three different thicknesses (i.e., 8 μm, 16 μm, or 24 μm). Since harbor porpoise teeth are spatulate in shape, two teeth were used for each individual and sectioned in two different orientations, one parallel to the mandible (porpoise cut) and one perpendicular (dolphin cut). Both cuts were made to ensure that the optimum sections were obtained. The other delphinids species (i.e., common dolphin, striped dolphin, Atlantic spotted dolphin, Atlantic white-sided dolphin, white-beaked dolphin, and bottlenose dolphin) had conical teeth and longitudinal tooth sections cut in the bucco-lingual plane or perpendicular to the mandible (dolphin cut).

The paraffin technique was adapted from Slooten (1991). In our study, decalcified teeth from 67 individuals were prepared using this technique, although the paraffin embedding procedure was modified at several stages (see details below). Decalcified teeth were rinsed in running tap water and bisected longitudinally using a microtome blade. Then the cut surface was placed face down into a plastic histology cassette for embedding in paraffin. The cassettes were processed in an automated machine for embedding the tissue in paraffin wax (Leica TP1050; Leica Microsystems GmbH, Wetzlar, Germany). During the process of paraffin embedding, tissues were dehydrated through increasing concentrations of alcohol, and then xylene and molten paraffin (see below for details). One of the technical problems that we found using the paraffin embedding procedure was that dolphin teeth were hardened and it was therefore difficult to obtain tooth sections using the standard microtome. Therefore, several trials were carried out to optimize the length of time that the teeth remained in ethanol, alcohol, and xylene to avoid excessive dehydration that consequently hardens the tissue.

The optimum program lasted around 3.5 h (as compared to 24 h for soft tissues). As it was mentioned above, tissues suffered a dehydration process passing through increasing concentrations of alcohol as follow: (1) 50% ethanol 30 min, (2) 70% ethanol 30 min 20°C, (3) 96% absolute alcohol 15 min at 20°C, (4) 100% absolute alcohol 15 min at 20°C, then (5) xylene 30 min at 20°C, (6) xylene 15 min at 20°C, and finally (7) molten wax (60°C) 30 min, and (8) molten wax (60°C) 15 min.

Finally the teeth were embedded in molten paraffin (60°C) using a blocking machine (Sakura Tissue-Tek TEC, Torrance, CA). They were then sectioned longitudinally at 8 μm thickness by using a standard paraffin microtome (Leica) with a stainless steel disposable microtome blade. The effect of section thickness was not investigated in the paraffin technique, because the sectioning equipment available did not allow preparation of thicker sections of high quality. Sections were oven-dried for 15 min at 100°C and the paraffin wax was then removed by passing the tooth sections through the following stages: (1) xylene 2 min, (2) xylene 2 min, (3) 100% alcohol 2 min, (4) 96% alcohol 2 min, (5) 70% alcohol 2 min, and (6) running distilled water before staining.

For 61 of these individuals, sufficient teeth were available to prepare stained sections using three different staining methods.

Staining and Mounting

For both techniques, multiple longitudinal tooth sections were obtained and only those most complete (i.e., including the crown and most of the pulp cavity) and closest to the midline were selected and stained using four different stains. These stains are commonly used for revealing the GLGs present in dentine and cementum in marine mammals: Mayer's hematoxylin (Thomas 1977, Myrick et al. 1983, Hohn and Lockyer 1995), Ehlrich's hematoxylin (Klevezal and Kleinenberg 1967, Korytin 1984), Toluidine blue (Thomas 1977, Graf and Wandeler 1982, Allen and Melfi 1985), and Giemsa (Stone et al. 1975; Matson 1981, 1993; Molina and Oporto 1993). Mayer's hematoxylin and Ehlrich's hematoxylin stained sections were “blued” in a weak ammonia solution and rinsed in distilled water (Hohn and Lockyer 1995), whereas Toluidine blue (Hohn and Lockyer 1995, Lockyer 1995) and Giemsa (Molina and Oporto 1993) stained sections were placed directly in distilled water before mounting.

Stained sections were mounted on microscope slides pre-coated with a 5% gelatin solution in order to prevent curling. Once the sections were fully dried on a warm hot plate, permanent slides were prepared using DPX-mountant.

Age Determination

Stained tooth sections were examined under a binocular microscope (×10–50 magnification). The age was estimated by counting the GLGs (Perrin and Myrick 1980) in the dentine and cementum (the latter only in those cases in which GLGs could be easily distinguished and/or those individuals which only one tooth from the same animal was available). We assumed that the GLGs represent one year's individual growth (although for the purpose of comparing techniques this assumption was not strictly necessary). Ages of teeth showing less than one full GLG were estimated to the nearest 0.5 GLG (6 mo).

The reading of the GLGs was carried out “blind” with no reference to biological data to avoid any possible biases in the estimation. Tooth sections were read three times by two independent readers (PLL and JAL). The final estimated age was achieved by consensus between the two readers. If no consensus was reached, the tooth was excluded from subsequent analysis.

Data Analysis

Quality of tooth preparations— We defined quality of the tooth preparation based on how clearly the GLGs could be identified as countable units. Thus, tooth preparations were categorized as follows: good quality (when GLGs were distinct, and easily counted), satisfactory (GLGs could be distinguished, but there was some uncertainty about how many GLGs were present), and poor quality (a reasonable estimate of the number of GLGs was impossible).

For the paraffin technique we examined the quality of tooth preparations stained using Mayer's hematoxylin, Ehlrich's hematoxylin, and Toluidine blue. The staining method which gave the highest percentage of good quality preparations was selected for examining the difference in quality between paraffin and cryostat techniques.

Variability in counting of GLGs— For the cryostat technique, we assessed the degree of variability associated with counts of GLGs for each particular staining method. Thus, we treated age estimates from each section thicknesses (i.e., 8, 16, and 24 μm) as individual observations and calculated the standard deviation values (SD) between them for each of the four stains. The overall variability for each particular stain is expressed using the frequency distribution of SD values.

To assess the variability between the two techniques, we treated estimates from each stain (i.e., Mayer's hematoxylin, Ehlrich's hematoxylin, and Toluidine blue) as observations and calculated the SD between them for the two techniques.

Statistical Analysis: The Effects of Several Factors on the Estimation of Age within Each Technique

Linear mixed modeling techniques (Pinheiro and Bates 2000, West et al. 2006, Zuur et al. 2007) were used to model estimated age in relation to several explanatory variables, namely, staining method, section thickness (μm), species, body length (cm), sex, and individual. Due to the nested structure of the data, species and dolphin identity (nested in species) were used as random effects.

An initial analysis indicated violation of homogeneity, and therefore we allowed for heterogeneous residual variance structures (Pinheiro and Bates 2000, West et al. 2006, Zuur et al. 2007). Thus, the following model was applied on the data obtained by the paraffin technique:

  • image

Ageijk is the age obtained by stain k for specimen j of species i, where k= 1, … , 3, i= 1, … , 6, and j takes any value between 3 and 25. The notation above means that age is modeled as a function of length, sex, and stain using the main terms, two-way interactions, and the three-way interaction. The terms ai and bi are random effects representing the between-species variation and between-specimen variation within a species. Both are assumed to be normally distributed with mean 0 and variances σ2a and σ2b, respectively. The term ɛijk is the unexplained noise (or within specimen variation), and is assumed to be normally distributed with mean 0 and variance given by

  • image

This variance structure allows for the modeling of heterogeneous residuals using a power function; see Pinheiro and Bates (2000) for the mathematical details. To avoid numerical problems, length was expressed in meters in the variance structure.

A similar model was applied to data obtained by the cryostat technique, in this case also including the effect of section thickness, and the comparison of data obtained by both main techniques.

The model selection followed the step-down approach described in West et al. (2006). All analyses were done in R (R Development Core Team 2006) using the nlme Package (Pinheiro et al. 2006).

Results

  1. Top of page
  2. Introduction
  3. Materials and Methods
  4. Results
  5. Discussion
  6. Acknowledgments
  7. Literature Cited

Quality of Tooth Preparation

Using the paraffin technique, the quality of tooth-preparations stained using the three staining methods, was examined for 61 individuals from several small cetacean species: harbor porpoise, common dolphin, striped dolphin, Atlantic spotted dolphin, and bottlenose dolphin, Atlantic white-sided dolphin, and white-beaked dolphin. However, poor quality preparations were obtained from pygmy sperm whale and dwarf sperm whale (Kogia sima) and results were not considered for further analyses because no further tooth samples from the same animal were available. Due to the lack of availability of more teeth for these individuals, it was not possible to repeat the procedure and they were excluded from further analyses.

Results indicated that the percentage of tooth preparations falling into each quality category varied substantially between stains (Fig. 1). Thus, using Mayer's hematoxylin, 70% of tooth preparations were considered “good quality,” 21%“satisfactory,” and 8%“poor quality.” Considerably lower proportions of good quality sections were obtained using Toluidine blue or Ehlrich's hematoxylin (51% and 31% of tooth preparations, respectively). Giemsa stained sections were considered to be of the poorest quality by both readers and were not included in any further comparative analysis since GLGs could not be distinguished.

image

Figure 1. Frequency distribution for quality of tooth preparations (n= 61) using the paraffin technique and three different staining methods (MH: Mayer's hematoxylin, EH: Ehlrich's hematoxylin, and TB: Toluidine blue). Tooth preparations were categorized using a three-grade scale as follows: poor quality, satisfactory quality, and good quality.

Download figure to PowerPoint

We examined the quality of tooth preparations processed by both techniques in 67 animals (all species combined). Based on Mayer's hematoxylin stained thin (8 μm) sections, the percentage of preparations considered of “good quality” was higher for the paraffin technique (37%) than for the cryostat technique (10%), although the percentage of “satisfactory” preparations was the same (46%) for both techniques (Fig. 2).

image

Figure 2. Frequency distribution for quality of Mayer's hematoxylin stained preparations using the paraffin and the cryostat techniques (n= 67). Tooth preparations were categorized using a three-grade scale as follows: poor quality, satisfactory quality, and good quality.

Download figure to PowerPoint

Variability in Counting of GLGs

Using the cryostat technique, the frequency distribution of the standard deviation values, (note: each individual SD value expressing variation in estimated ages from different section thicknesses for a single tooth), was similar for the four staining methods. For most individuals, differences between age estimates obtained using different section thicknesses were small, as indicated by SD values that range from 0 to 1.5 (Fig. 3). The absence of any high SD values suggested that variability of GLG counts was slightly lower for Mayer's hematoxylin than other staining methods.

image

Figure 3. Frequency distribution of values for SD (expressing variation in estimated ages between the three section thicknesses) for the four staining methods (i.e., MH: Mayer's hematoxylin, EH: Ehlrich's hematoxylin, TB: Toluidine blue, G: Giemsa) (n= 30 in each case).

Download figure to PowerPoint

The frequency distributions of the SD values expressing variation in estimated ages from different stains (namely, Mayer's hematoxylin, Ehlrich's hematoxylin, and Toluidine blue) were similar for both techniques. For most individuals, differences between age estimates were small, as was indicated by SD values that range from 0 to 1.5 (Fig. 4). The variability of GLG counts thus appeared to be low and similar for both techniques.

image

Figure 4. Frequency distribution of values for SD (expressing variation in estimated ages between three staining methods, i.e., MH: Mayer's hematoxylin, EH: Ehlrich's hematoxylin, TB: Toluidine blue) for the two techniques (n= 34).

Download figure to PowerPoint

Paraffin Technique

The optimal model for estimated age using the paraffin data contained the main terms stain and body length, and the interaction between them. Estimated parameters are given in Table 2. Using a likelihood ratio test, a model with, and a model without the interaction term were compared, and the results indicated that the interaction term was significant (L= 25.420, df = 2, P < 0.001).

Table 2.  Estimated parameters for the paraffin data. The term stain was fitted as a nominal variable and length as a continuous variable. The random effect ai representing the between-species variation is N (0, 6.702) and the random effect bij, representing the between-animal variation in the same species is N (0, 3.052). The estimated values for σ and δ are 0.23 and 2.21, respectively.
 ValueSEdft-valueP-value
(Intercept)−23.17233.8571114−6.00760
factor (Stain: Ehlrichs hematoxylin)1.86870.40301144.63680
factor (Stain: Toluidine blue)1.14140.40301142.83200.0055
Length0.17370.01555111.13560
factor (Stain: Ehlrichs hematoxylin):Length−0.01780.0032114−5.54830
factor (Stain: Toluidine blue): Length−0.00970.0032114−3.03660.0030

Ehlrich's hematoxylin and Toluidine blue stains were different from the baseline (Mayer's hematoxylin), and the age–length relationship differed between staining methods.

Cryostat Technique

Similar results were obtained for the cryostat technique data. Thus, the optimal model contained the main terms stain and body length, and the interaction between them (see Table 3 for the estimated parameters). Using a likelihood ratio test, a model with, and a model without the interaction term were compared, and the results indicated that the interaction term was significant (L= 10.079, df = 3, P= 0.0179).

Table 3.  Estimated parameters for the cryostat data. The term stain was fitted as a nominal variable and length as a continuous variable. The random effect ai representing the between-species variation is N (0, 6.292) and the random effect bij, representing the between-animal variation in the same species is N (0, 3.732). The estimated values for σ and δ are 0.096 and 3.798.
 ValueSEdft-valueP-value
(Intercept)−17.074.3406313−3.93310
factor (Stain: Ehlrich's hematoxylin)1.01830.34863132.92110.004
factor (Stain: Toluidine blue)0.55850.34863131.6020.1102
factor (Stain: Giemsa)0.68140.34863131.95480.0515
Length0.13620.0204206.65730
factor (Stain Ehlrich's hematoxylin): Length−0.010.0028313−3.12680.002
factor (Stain: Toluidine):Length00.0028313−1.73280.0841
factor (Stain: Giemsa):Length−0.010.0028313−2.09110.0373

Ehlrich's hematoxylin and Giemsa stains were different from the baseline (Mayer's hematoxylin), giving higher ages (as indicated by positive coefficients) and the apparent age–length relationship differed between stains, as indicated by the significant interaction terms. There was no effect of section thickness.

Comparison of Both Techniques

The optimal model for the comparison of both techniques data contained only body length as the main term. There was no difference between techniques. Estimated parameters are given in Table 4. There was more variation in the age readings of larger animals.

Table 4.  Estimated parameters for the comparison. The term technique was fitted as a nominal variable and length as a continuous variable. The random effect ai representing the between-species variation is N (0, 4.7082) and the random effect bij, representing the between-animal variation in the same species is N (0, 2.2952). The estimated values for σ and δ are 0.156 and 3.822.
 ValueSEdft-valueP-value
(Intercept)−23.912.907365−8.22430
Length0.17370.01285613.56220

Discussion

  1. Top of page
  2. Introduction
  3. Materials and Methods
  4. Results
  5. Discussion
  6. Acknowledgments
  7. Literature Cited

Quality of Tooth Preparations

Results from the paraffin technique showed that the best contrast of GLGs was obtained using Mayer's hematoxylin which gave the highest percentage of tooth preparations classified as good quality. Use of Ehlrich's hematoxylin did not improve and rather decreased the contrast of the growth layers in comparison to Mayer's hematoxylin and Toluidine blue. Giemsa was the staining method that was least successful in terms of revealing growth layer patterns. The Giemsa stained preparations were considered the “poorest” quality by both readers and were not included in the comparative analysis, because GLGs could not be identified. In contrast, Giemsa-stained sections obtained using the cryostat technique showed a reliable contrast of the growth layers.

This suggests that the effectiveness of Giemsa in revealing dentinal growth layers might depend on the procedure for preparing teeth before staining. It is possible that the procedure of drying and passing the slides through a standard alcohol series in order to remove the paraffin affects the penetration of Giemsa stain into the tissue, giving different interpretations of growth layers and therefore differences in estimated ages. Giemsa stain has been used for age determination from cementum layers (Matson 1981, 1993) and has been recommended by some authors (Stone et al. 1975). However, no published information is available about the application of this method for revealing dentinal growth layers or even its application on dolphin teeth. Results of this study indicated that further experimental work is required, for example, to investigate the main components of the dentine and cementum tissues (i.e., the organic and inorganic material) for which Giemsa stain has more affinity. We also examined the quality of tooth preparations processed by both techniques based on Mayer's hematoxylin-stained thin sections. Results showed that the percentage of preparations considered of “good quality” was higher for the paraffin technique than for the cryostat technique. This might suggest that the standard microtome is better for obtaining thin sections than the freezing microtome, particularly when working with small teeth. Nevertheless, most preparations showed an acceptable (satisfactory) layering pattern regardless of technique. This indicated that the paraffin technique can be applied as a viable tooth preparation technique, being especially successful for preparing small teeth (e.g., harbor porpoises), for which thin sections are recommended to ensure multiple replicates from the same tooth close to the pulp cavity (along the midline). In addition, the paraffin technique was successfully applied to a reasonably wide range of small cetacean species including those with small teeth such as spotted dolphin, common dolphin, and Atlantic white-sided dolphin, and other species with larger teeth including bottlenose dolphin and pygmy sperm whale. However, the paraffin technique did not seem to work well for the dwarf sperm whale, suggesting that it may be more appropriate for preparing teeth of delphinids.

Variability in Counting of GLGs

Results indicated that, using the cryostat technique, differences between estimates from different section thicknesses were small and similar for the four staining methods. However, the variability of GLG counts was lower for Mayer's hematoxylin than other staining methods. As was mentioned above, regarding the quality of tooth preparations when using the paraffin technique, the highest percentage of tooth preparations catalogued of good quality was obtained using Mayer's hematoxylin. This might indicate that Mayer's hematoxylin provides a reliable contrast of the growth layers regardless of the prior preparation technique.

Using both techniques, differences between estimates from three staining methods (i.e., Mayer's hematoxylin, Ehlrich's hematoxylin, and Toluidine blue) were also small. The variability in GLG counts appeared to be similar for both techniques. Again, regardless of the staining method, the paraffin technique can be considered a reliable tooth preparation technique.

The Effect of Stain on Estimated Ages

Results showed that the stain contributed significantly to the estimated ages using either the paraffin or the cryostat technique.

Other aspects of the procedure for preparing teeth (e.g., decalcification, sectioning, and mounting) can affect the ease of distinguishing growth layers. For example, some researchers (Graf and Wandeler 1982, Allen and Melfin 1985) have preferred Toluidine blue to hematoxylins, but they suggested that it is necessary either to mount sections in synthetic resins or to study a preparation just after it has been made, because the type of stain (i.e., Toluidine blue) disappears in glycerine. Korytin (1984) tested four hematoxylins for staining teeth and bone tissue. He concluded that the best contrast of layers was obtained by using Krutsai hematoxylin with an increased amount of hematoxylin powder and decreased amount of HCL. Thomas (1977) tried 20 different stains, besides Ehrlich's and Harris's hematoxylins, for staining cementum layers in carnivores and ungulates. He concluded that the order of preference of the stains in revealing cementum layers can be different for different species. Therefore, other factors such as the recipe for the stain, time of staining, age of the stain as well as inter- and intraspecific differences in the layering pattern may also affect the visualization of GLGs and therefore the interpretation of age.

Using the paraffin technique, Ehlrich's hematoxylin and Toluidine blue stains were different from the baseline (i.e., Mayer's hematoxylin), giving higher ages as indicated by positive coefficients. Similar results were obtained from the cryostat technique data, in which Ehlrich's hematoxylin and Giemsa stains were different from Mayer's hematoxylin. Results of this study regarding the quality of stained tooth sections suggest that Ehlrich's hematoxylin shows the GLGs less reliably in comparison to other stains.

Furthermore, in both techniques, the apparent age–length relationship differs between stains, as indicated by the significant interaction terms. Overall, the age was probably underestimated for older adults in which the pulp cavity was almost closed. Bryden (1989) pointed out that one of the problems encountered in age determination by counting growth layers in cetacean teeth, particularly those from small cetacean species, like harbor porpoises, is that teeth are small and dentinal deposition may occlude the pulp cavity before the animal dies, giving only minimum age estimates for older adults. When the tooth grows, dentinal growth layers are deposited inwards until the pulp cavity is occluded. Thus layers become narrower and it is more difficult to identify the boundary of the most recently deposited growth layer, potentially leading to incorrect interpretation of age. Therefore an underestimation of age is expected for older adults due to the occlusion of the pulp cavity regardless of the tooth preparation technique used.

The Effect of Section Thickness

Using the cryostat technique, teeth were sectioned at 8, 16, and 24 μm thickness. Results showed that section thickness did not contribute significantly on the estimation of age. Using the paraffin technique, teeth were sectioned at 8 μm thickness. Particularly when working with small dolphin teeth, this can be considered an advantage since this guarantees a suitable set of central section's close to the pulp cavity, which is recommended in order to ensure a reliable interpretation of the layering pattern and an accurate estimation of age. The ability to obtain thin sections might provide an additional advantage in those cases in which sample sizes are small, therefore guaranteeing a higher number of multiple sections from the same tooth.

In cetacean species with larger teeth, like bottlenose dolphins, thicker tooth sections are recommended (Perrin and Myrick 1980). However, with the paraffin embedding procedure it is very difficult to obtain thick sections using the standard microtome, because the process of embedding hardens the tissue. In a recent study, Duignan and Jones (2005) found some technical problems with sectioning when they tried to obtain sections of 10–20 μm thickness from Hector's dolphin teeth. The tooth sections were of poor quality and age was not estimated with certainty. In our study, we solved some of the technical problems that we found with sectioning, although we were not able to obtain sections thicker than 8 μm. However, results showed that thin sections provided a reliable estimated age for all studied species, even for those with larger teeth.

Finally, we investigated whether the technique (i.e., paraffin or cryostat) had a significant effect on the estimated ages. The absence of significant differences between estimated ages from both techniques indicated that the paraffin technique can be applied as a tooth preparation technique alternative to the cryostat technique.

As no teeth from known age animals were available, we were not able to show which technique was more efficient in obtaining accurate counts. Ideally, similar work is required using teeth from known age animals to explore whether estimated age corresponds to the real age.

Concluding Remarks

We conclude that the paraffin technique represents a viable and cost-effective alternative to use of a cryostat or freezing microtome when preparing cetacean teeth for age determination.

However:

  • 1
    The effect of the staining method on the estimated age should be taken into account in both techniques.
  • 2
    Using the paraffin technique, Ehlrich's hematoxylin and Toluidine blue stains are different from the baseline (Mayer's hematoxylin), and the apparent age–length relationship differs between staining methods. However, the variability of GLG counts is small and appears to be similar for the three staining methods.
  • 3
    Using the cryostat technique, the variability of GLG counts is lower for Mayer's hematoxylin than other staining methods. Ehlrich's hematoxylin and Giemsa stains are different from the baseline (Mayer's hematoxylin), giving higher ages and the apparent age–length relationship also differs between stains.
  • 4
    We recommend use of longitudinal tooth sections of 8 μm, particularly in small cetacean species, and when the availability of tooth samples is limited.
  • 5
    When using the paraffin technique, we recommend use of Mayer's hematoxylin for staining, which gave the highest percentage of preparations classified as a “good quality.”
  • 6
    Overall the paraffin technique requires less specialized equipment than the cryostat technique.
Footnotes
  • 1

    Personal communication from P. J. Duignan, Institute of Natural Resources, Massey University, Albany Campus, The Coastal Marine Research Group, Private Bag 102 904, North Shore MSC, Auckland, New Zealand, 2006.

Acknowledgments

  1. Top of page
  2. Introduction
  3. Materials and Methods
  4. Results
  5. Discussion
  6. Acknowledgments
  7. Literature Cited

We are grateful to Bob Reid from the Scottish Agricultural College (SAC) Veterinary Services at Inverness as well Manolo Arbelo and colleagues from the Unit of Histology and Veterinary Pathology, Institute for Animal Health, Veterinary School, University of Las Palmas de Gran Canaria (Spain) for the collection of all material used for this study. We are also very grateful to Dr. Michel André and Dr. Eduardo Degollada for their advice and help with the research. This study was funded by a Marie Curie grant to PLL, as part of the Marie Curie training site project QLK5-1999-50530, supported by the European Commission. We thank Bill Edwards (School of Biological Sciences, University of Aberdeen) for his support and advice, and the Pathology Department in the Medical School (University of Aberdeen) for access to facilities and technical support. GJP would also like to acknowledge support from the ANIMATE project (MEXC-CT-2006-042337).

Literature Cited

  1. Top of page
  2. Introduction
  3. Materials and Methods
  4. Results
  5. Discussion
  6. Acknowledgments
  7. Literature Cited
  • Allen, D. S., and R. C. Melfi. 1985. Improvements in techniques for ageing mammals by dental cementum annuli. The Proceedings of the Iowa Academic of Science 93:100102.
  • Bryden, M. M. 1989. Age and growth. Pages 211224 in M. M.Bryden and R.Harrison, eds. Research on dolphins. Claredon Press, Oxford , UK .
  • Collet, A. 1981. Biologie du dauphin commun Delphinus delphis (L.) en Atlantique Nord-Est. Ph.D. thesis, L’Université de Poitiers, Poitiers , France . 156 pp.
  • Duignan, P. J., and G. W. Jones. 2005. Autopsy of cetaceans including those incidentally caught in commercial fisheries, 2002/03. DOC Science Internal Series 195. Department of Conservation, Wellington , New Zealand . 22 pp.
  • Graf, M., and A. I. Wandeler. 1982. Alterbestimmung bei Dachsen (Meles meles L.). Revue Suisse de Zoologie 89:10171023.
  • Gurevich, V. S., B. S. Stewart and L. H. Cornell. 1980. The use of tetracycline in age determination of common dolphins, Delphinus delphis. Report of the International Whaling Commission (Special Issue 3):165169.
  • Hohn, A. A. 1980a. Analysis of growth layers in the teeth of Tursiops truncatus using light microscopy, microradiography, and SEM. Report of the International Whaling Commission (Special Issue 3):155160.
  • Hohn, A. A. 1980b. Age determination and age related factors in the teeth of Western North Atlantic bottlenose dolphins. Scientific Reports of the Whales Research Institute, Tokyo 32:3966.
  • Hohn, A. A. 1990. Reading between the lines: An analysis of age estimation in dolphins. Pages 575585 in S.Leatherwood and R. R.Reeves, eds. The bottlenose dolphin. Academic Press, New York , NY .
  • Hohn, A. A. 2002. Age estimation. Pages 613 in W. F.Perrin, B.Würsig and J. G. M.Thewissen, eds. Encyclopedia of marine mammals. Academic Press, San Diego , CA .
  • Hohn, A. A., and S. Fernández. 1999. Biases in dolphin age structure due to age estimation technique. Marine Mammal Science 15:11241132.
  • Hohn, A. A., and C. H. Lockyer. 1995. Report of the harbour porpoise age determination workshop, Oslo, 21–23 May 1990. Appendix 3: Protocol for obtaining age estimates from harbour porpoise teeth. Report of the International Whaling Commission (Special Issue 16):494496.
  • Hohn, A. A., and C. Pinedo. 2000. Growth layer pattern in teeth from the Franciscana (Pontoporia blainvillei): Developing a model for precision in age estimation. Marine Mammal Science 16:127.
  • Hohn, A. A., M. D. Scott, R. S. Wells, J. S. Sweeney and A. B. Irvine. 1989. Growth layers in the teeth from known-age, free-ranging bottlenose dolphins. Marine Mammal Science 5:315342.
  • Humason, G. L. 1962. Animal tissue techniques. W. H. Freeman & Co., San Francisco , CA .
  • Klevezal, G. A. 1980. Layers in the hard tissues of mammals as a record of growth rhythms of individuals. Report of the International Whaling Commission (Special Issue 3):8994.
  • Klevezal, G. A. 1996. Recording structures of mammals: Determination of age and reconstruction of life history. A. A. Balkema, Rotterdam , The Netherlands .
  • Klevezal, G. A., and S. E. Kleinenberg. 1967. Age determination of mammals by layered structures of teeth and bones. M.: Nauka. Translated 1969 from Russian by Israel Program for Scientific Translations.
  • Korytin, N. S. 1984. [Testing four different stains for their efficiency to stain sections of teeth and bones]. In Russian. Registriruyushchie strucktury I opredelenie vozrasta mlekopitayushchikh. Vsesoyuznaya Konf. 22–25 May, Moscow . Tezisy Doklavov:3132.
  • Kuiken, T., and M. G. Hartmann. 1991. Cetacean pathology: Dissection techniques and tissue sampling. Proceedings of the first European Cetacean Society workshop on cetacean pathology: Dissection techniques and tissue sampling, Leiden, The Netherlands. European Cetacean Society Newsletter no. 17.
  • Lockyer, C. H. 1993. A report on patterns of deposition of dentine and cement in teeth of pilot whales, genus Globicephala. Report of the International Whaling Commission (Special Issue 14):137162.
  • Lockyer, C. H. 1995. A review of factors involved in zonation of Odontocete teeth, and an investigation of the likely impact of environmental factors and major life events on harbour porpoise tooth structure. Report of the International Whaling Commission (Special Issue 16):511529.
  • Matson, G. M. 1981. Workbook for cementum analysis. Matson's Laboratory, Milltown , MT .
  • Matson, G. M. 1993. Progress report 13. Matson's Laboratory, Milltown , MT .
  • Molina, D. M., and J. A. Oporto. 1993. Comparative study of dentine staining techniques to estimate age in the Chilean dolphin (Cephalorhynchus eutropia, Gray, 1846). Aquatic Mammals 19:4548.
  • Myrick, A. C., and L. H. Cornell. 1990. Calibrating dental layers in captive bottlenose dolphins from serial tetracycline labels and tooth extractions. Pages 587608 in S.Leatherwood and R. R.Reeves, eds. The bottlenose dolphin. Academic Press, San Diego , CA .
  • Myrick, A. C., A. A. Hohn, P. A. Sloan, M. Kimura and D. D. Stanley. 1983. Estimating the age of spotted and spinner dolphins (Stenella attenuata and Stenella longirostris) from teeth. National Oceanic Atmospheric Administration Technical Memorandum NMFS, SWFC-30. 17 pp.
  • Myrick, A. C., E. W. Shallenberger, I. Kang and D. B. Mackay. 1984. Calibration of dental layers in seven captive Hawaiian spinner dolphin Stenella longirostris, based on tetracycline labelling. Fisheries Bulletin 82:207225.
  • Nielsen, H. G. 1972. Age determination of the harbour porpoise Phocoena phocoena (Cetacea). Videnskabelige Meddelelser fra dansk Naturhistoric Forening 135:6164.
  • Perrin, W. F., and A. C. Myrick. 1980. Age determination of toothed whales and sirenians. Report of the International Whaling Commission (Special Issue 3):150.
  • Perrin, W. F., D. B. Holts and R. B. Miller. 1977. Growth and reproduction of the eastern spinner dolphin, a geographical form of Stenella longirostris in the Eastern Tropical Pacific. Fishery bulletin 75:725750.
  • Pinheiro, J. C., and D. M. Bates. 2000. Mixed-effects models in S and S-Plus. Springer-Verlag, New York , NY .
  • Pinheiro, J. C., D. M. Bates, S. Debroy and D. Sarkar. 2006. NLME: Linear and nonlinear mixed effects models. R package version 3.
  • R Development Core Team. 2006. R: A language and environment for statistical computing. R Foundation for Statistical Computing, Vienna , Austria . http://www.R-project.org.
  • Slooten, E. 1991. Age, growth and reproduction in Hertor's dolphins (Cephalorhynchus hectori). Canadian Journal of Zoology 69:16891700.
  • Stone, W. B., A. S. Clauson, D. E. Slingerlands and B. L. Weber. 1975. Use of Romanowsky stains to prepare tooth sections for aging animals. New York Fish and Game Journal 22:156158.
  • Thomas, D. C. 1977. Metachromatic staining of dental cementum for mammalian age determination. Journal of Wildlife Management 41:207210.
  • West, B., K. B. Welch and A. Galecki. 2006. Linear mixed models: A practical guide using statistical software. Chapman and Hall/CRC Press, Boca Raton , FL .
  • Zuur, A. F., E. N. Ieno and G. M. Smith. 2007. Analyzing ecological data. Springer, New York , NY .