The Chemistry of Beeswax

Abstract

Publications on the physical constants for the comb waxes of Asian and European beeswaxes first appeared a century ago. It was soon shown that carbon chain length was, on average, shorter in the Asian beeswaxes than in A. mellifera, which explains the lower melting points of the former. The Asian waxes are more similar to one another than to A. mellifera. In Asian beeswaxes, the amounts of C₃₁ and C₃₃ in the pool of free fatty acids are reduced, but C₂₅ hydrocarbons are increased compared to that of A. mellifera. The major compound families in beeswax are alkanes, alkenes, free fatty acids, monoesters, diesters and hydroxymonoesters, while fatty alcohols and hydroxydiesters are minor constituents. There are notable species-specific differences in the beeswaxes among honeybee species, but all share a complex mixture of homologous neutral lipids. The amounts of acylglycerols are the same in scale and comb wax, but diacylglycerols dominate the former and monoacylglycerols the latter. There are more double-bonded fatty acids in comb than in scale wax, and a greater saturation of fatty acids in comb wax. Beeswaxes analysed with high temperature gas chromatography yielded a characteristic elution pattern for waxes of each honeybee species. A parsimonious, unweighted, pair-group analysis based on the distribution of the chemical constituents for 82 elution peaks of the derivatized comb waxes of six species of honeybees. The Euclidean distances of the beeswaxes present a picture very similar to those obtained from morphometric, behavioural and DNA sequence analyses. The wax glands and the products of their secretions were highly conserved features during honeybee evolution.

These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.

16.1 Introduction

In this chapter, discussions on the chemistry of beeswax are restricted entirely to honeybee wax scales and comb wax in a biological context. Investigations of both the chemical composition and physical properties of beeswaxes of A. mellifera have been pursued for centuries, and these earlier works have been documented by Grün and Halden (1929). Preparations for and practical uses of beeswax have also been documented (Cowan 1908; Coggshall and Morse 1995), and the commercial industrial aspects of beeswax have been exhaustively monographed (Büll 1977); thousands of publications have appeared on these topics since then. However, the very first studies of Asian beeswaxes appeared only a century ago (Hooper 1904; Bellier 1906; Büchner 1906; Hooper and Büchner 1906; Ueno 1915; Roberts and Islip 1922; Ikuta 1931, 1934), who between them recorded the physical constants (specific gravity, melting point, acid and saponification values , etc.) of the comb waxes of A. cerana, A. dorsata, A. florea and A. mellifera.

As our knowledge of the hydrocarbon, alcohol and acid fractions of beeswaxes developed, two points of importance to honeybee biology emerged. Firstly, Phadke (1961) re-examined the physical constants of A. cerana, A. dorsata, A. florea and A. mellifera beeswaxes, and showed each to be extremely homogenous as evidenced by the very small standard deviations in the physical values of the samples measured. Shortly after, Narayana (1970) and Phadke et al. (1971) determined that carbon chain length was, on average, shorter in the three Asian beeswaxes than in A. mellifera, which accounts for the lower melting points of the Asian waxes. Progress in wax chemistry advanced with gradually improved analytical techniques of both thin-layer and gas-liquid methods of chromatography in the 1940 and 1950s (Touchstone 1993).

16.2 Chemical Composition

The composition and origin of A. mellifera comb beeswax has relatively recently been summarised by Tulloch (1980), and is shown in Table 16.1. The major components are defined as those exceeding more than 5 % of each fraction; those of lesser abundance are regarded as minor constituents. Tulloch regarded, as major components, those which constituted more than 1 % of each fraction; those of lesser abundance were regarded as minor constituents. Nevertheless, if a particular fraction is itself small, then a given compound may well be ‘major’ in that fraction, but very minor with respect to the bulk composition of a beeswax sample. Tulloch (1980) regarded the large number of minor hydrocarbons as probably disproportionate, because of the relative ease with which they can be separated, vis-à-vis the seven groups of esters. The residue of some 44 % of beeswax is taken up entirely by minor constituents, to which Tulloch ascribed the relatively low melting point of intact beeswax and its plasticity.

Table 16.1 Composition of beeswax derived from A. mellifera combs (Tulloch 1980)^a

By combining both gasliquid and thin-layer methods of chromatography Tulloch (1973, 1974, 1975, 1980) also studied the composition of waxes from different honeybee species. He found that the waxes from different A. mellifera races were very similar as a group, but the unsaturated C₃₁ hydrocarbon peak was smaller and the C₃₅ hydrocarbon peak larger in the African bee, A. m. scutellata, than in the European races of A. mellifera. By contrast, he reported that waxes of the Asian bees, A. cerana, A. dorsata and A. florea, resemble each other more closely than any of them do to A. mellifera waxes as previously reported by Narayana (1970) and Phadke et al. (1971). In the Asian waxes there is a smaller pool of free fatty acids (analysed as methyl esters), reduced amounts of C₃₁ and C₃₃, but increased C₂₅ hydrocarbons compared to A. mellifera waxes. The recordings from the gas-liquid chromatography analyses by Tulloch are shown in Fig. 16.1.

Fig. 16.1

The spectra obtained from gas–liquid chromatographic analyses of A. mellifera, A. m. scutellata (= adansonii), A. dorsata, A. cerana and A. florea comb waxes. Hydrocarbons are indicated by odd-numbered peaks (23–35), free acids by even-numbered peaks (24–34) and monoesters by even numbers (40–50) (Tulloch 1980)

Despite the assiduous efforts of numerous chemists who have sought to analyse the composition of beeswax, we have very few observations on the chemistry of newly secreted wax scales. Huber (1814) investigated the solubility properties of wax scales and of fragments of newly fashioned white comb wax. He observed that the wax scales readily dissolved in turpentine (presumably comprising then, as now, a pot-pourri of terpenes, but mainly the monoterpenes ά- and β-pinene ), but that comb wax left a white residue. When scale and comb wax samples of equivalent weight were placed in vessels of sulphuric ether (probably diethyl ether), the former became opaque but did not dissolve, while the latter dissolved leaving a white residue in the vessel.

When Huber allowed the ether to evaporate from the vessels, he always obtained a recoverable layer of scale wax residue, which led him to conclude that if the scales were indeed crude wax, then the bees must impregnate them with some additional substance to obtain the whiteness and ductility of newly constructed comb wax. To this we can add the observations of Young (1963), who analysed wax scales for the presence of (2-¹⁴C)-acetate that had been injected into wax-producing bees. He found that the label was incorporated in the free acid and ester fractions of wax scales. Finally, Lambremont and Wykle (1979) performed a thin-layer chromatographic separation of scale wax and found the resulting chromatographic pattern similar to that obtained by Tulloch (1970) from cappings wax, with the exception that their chromatograms lacked activity at the diester position.

Subsequently, Davidson and Hepburn (1986) compared the glycerols of scale and comb wax. Their assays showed that the monoacylglycerol and diacylglycerol fractions comprised about 91 % of the total glycerol in scale and comb wax. While the total level of the acylglycerols were the same in scale and comb wax, the diacylglycerols dominated the scale wax glycerol pool, and the monoacylglycerols the comb wax glycerols. Within the acylglycerol fractions there were substantially more double-bonded fatty acids in scale than in comb wax. Although about 50 % of the fatty acid fractions were the same in the two waxes, there was a significantly greater degree of saturation in the fatty acids of comb wax.

In the absence of hard analytical knowledge as to the total composition of beeswax scales vis-à-vis that of newly built comb, a rather circuitous route must be taken to assess the possible differences among European, African and ‘africanized’ (A. m. scutellata) subspecies. Of equal importance, what exactly is it that a honeybee worker does when she chews scales in the process of comb construction? To this end Eckert (1922, 1927) repeated the basic experiment of Dumas and Edwards (1843) to assess the effects of cane sugar versus honey on the composition of wax. He compared the fresh, white wax of newly constructed combs built by bees fed sugar, with the yellowish wax produced by a colony given nectar and honey, and found no differences between them. The dimension of age was added to composition studies by Jordan et al. (1940), who compared old comb wax, wax newly secreted by young bees and new wax produced by bees of more than a month old. Replicate and parallel measurements were made on cleaned combs, but no significant differences were found between the waxes of young and old bees. These two waxes did, however, differ from old comb wax in that the latter had an iodine number twice that of the former. This they attributed to a greater contamination of the old wax by carotenoids derived from pollen.

16.3 Chemometrics

Titschack (1969) analysed and tabulated the acid, saponification and ester values for A. mellifera African waxes, ranging in origin from Morocco and Ethiopia through the Ivory Coast and south to Mozambique. Because these data were sorted by countries, individual results cannot confidently be ascribed to any particular honeybee subspecies (Hepburn and Radloff 1998). Nonetheless, there were statistically significant differences in composition between several African waxes from different sources, pointing to possible genetic differences among the races. This approach was extended by Tulloch (1980) who showed that the waxes of Asian honeybees were chemically different from those of A. mellifera, and that the African and European subspecific profiles of A. mellifera waxes also differed.

With the development of high resolution capillary gas chromatography, this work has been extended, particularly by Brand-Garnys and Sprenger (1988). They characterised the waxes of different A. mellifera races on the basis of unique hydrocarbon and ester profiles, and recognised 16 subspecific waxes, ten of which are of African geographical origin (Table 16.2). Unfortunately no information is given as to the origin of these waxes, or of variations between the samples, so these data elude chemotaxonomic analysis. Recently, Beverly et al. (1995) showed that the pyrolysis-mass spectral peaks obtained from European and African beeswaxes differed in their relative intensities, but no unique molecules peculiar to any specific wax were obtained. Nonetheless, this approach might be a useful line of further inquiry.

Table 16.2 Wax characteristics of different A. mellifera races (Brand-Garnys and Sprenger 1988)^a

With even more sophisticated gas-chromatographic methods than previously available Aichholz and Lorbeer (1999) and Aichholz et al. (2000) re-examined the comb waxes of the Asian honeybees, A. andreniformis, A. cerana, A. dorsata, A. florea and A. laboriosa as well as A. mellifera, and showed that they are complex mixtures of homologous neutral lipids containing a range of 20–64 carbon length molecules. Aichholz et al. (2000) investigated beeswaxes with high temperature gas chromatography and obtained a characteristic elution pattern for the waxes of each honeybee species, confirming and extending the earlier analyses of Tulloch (1980) and Brand-Garnys and Sprenger (1988).

In another analysis of beeswaxes Puleo (1991) published gas chromatograms of the comb waxes of African A. m. scutellata and European A. m. ligustica honeybees, and demonstrated striking differences in both their hydrocarbon and straight chain monoester fractions. In the former, the percentage of C₃₃:1 unsaturated hydrocarbon is greater than the concentrations of C₂₉ and C₃₁ saturated hydrocarbons, while the converse occurs in the latter subspecies. Also, the percentage of C35:1 unsaturated hydrocarbon is ten times greater in A. m. scutellata (~1.2) than in A. m. ligustica (~0.2). Likewise, there is a lower percentage concentration of C₄₈ relative to the C₄₆ esters in A. m. scutellata than in A. m. ligustica (Puleo 1991). He also reported that there are also minor components associated with the hydrocarbon fraction, in that the even-numbered, straight chain hydrocarbons vary in length from C₂₂ to C₃₄ and may constitute 0.02–0.2 % of the total.

Following Tulloch (1980), Aichholz et al. (2000) defined the major compound families as those exceeding 5 % of the total, so that alkanes, alkenes, free fatty acids, monoesters, diesters and hydroxymonoesters are the major compound families, while fatty alcohols and hydroxydiesters are minor constituents (Table 16.3). There are notable species-specific differences in the waxes among honeybee species (Table 16.3), but all share a complex mixture of homologous neutral lipids: C₂₅–C₂₉ alkanes, C₄₀–C₅₄ monoesters, C₄₂–C₅₂ hydroxymonoesters, and C₅₆–C₅₈ diesters (Aichholz and Lorbeer 1999; Aichholz et al. 2000). Presently our knowledge of the composition of the waxes of all honeybee species is nearly equal; however, pathways of synthesis remain available only for A. mellifera (Hepburn et al. 1991). Given what is known of species-specific composition (Table 16.4), there is considerable opportunity for biochemical studies of beeswaxes in future.

Table 16.3 The major compound families of A. andreniformis, A. florea, A. cerana, A. mellifera, A. dorsata and A. laboriosa comb waxes (Aichholz and Lorbeer 1999)

Table 16.4 Comparison of the compound composition of derivatised comb waxes of A. mellifera, A. cerana, A. florea, A. andreniformis, A. dorsata and A. laboriosa by GC–FID analysis on a SOP-50-PFD column (modified from Aichholz and Lorbeer 1999)

16.3.1 Chemometric Classification of Beeswaxes

For any experimental study into the numerous interactions between pheromones and comb and/or cuticular waxes known to occur (Breed et al. 1995a, b, 1998), it is essential to know the chemical composition of the waxes involved and to be able to classify them. The chemical compositions of comb and cuticular waxes of honeybees have been extensively investigated (Blomquist and Ries 1979; Blomquist et al. 1980; Lockey 1985; Hepburn 1986; Francis et al. 1989), but with different methods. In a seminal paper, Frölich et al. (2000) established objective and quantitative chemometric tools for distinguishing between comb waxes of different ages and the cuticular waxes from different castes and sexes of A. m. carnica. Previously there had been no studies on chemical composition of different age classes of comb waxes using quantitative classification tools.

When Frölich et al. (2000) analyzed their fractions by gas chromatography, 56–75 % of the total mass of the wax samples could be identified (Table 16.5). All comb waxes of different age classes were dominated by long-chain aliphatic compounds, with chain lengths ranging in length from C₂₁ to C₅₄ (Fig. 16.2). The chain lengths exhibited a bimodal distribution, and there were no differences in chain length distributions among wax scales, new, middle-aged, and old comb waxes respectively. The respective medians for the shorter and longer chain length distributions were also fairly close. Chain lengths were in the range of C₄₂ to C₄₄ for all comb wax classes (Fig. 16.2). These data are consistent with those of other studies on A. mellifera (Basson and Reynhardt 1988), as well as waxes of the Asian honeybee species (Narayana 1970; Phadke et al. 1971).

Table 16.5 Analytical yields derived from gas chromatographic analyses of A. m. carnica scale and comb waxes (Frölich et al. 2000)

Fig. 16.2

Distribution of chain lengths of A. m. carnica comb waxes. Median₁ refers to the chains ranging from C₁₉ to C₃₆; Median₂ refers to the chains ranging from C₃₇ to C₅₄; and Median_all characterizes the whole range of chain lengths (Frölich et al. 2000)

The chemical compositions of all waxes were dominated by long-chain alkyl esters contributing 47 % ± 4.0 to 57 % ± 6.9 of the total of fractions 1–3 (Table 16.6).

Table 16.6 Relative chemical composition of A. m. carnica comb waxes of different ages (Frölich et al. 2000)

With the increasing age of comb wax, the overall median of the different age classes decreases, but the relative contributions by alkenes, alkadienes and branched alkanes increased from 3.4 % ± 1.43 (alkenes), 0.06 % ± 0.044 (alkadienes) and 0.00 % ± 0.008 (branched alkanes) in wax scales, to 12 % ± 1.3, 2.0 % ± 0.21 and 0.95 % ± 0.129 in old comb wax respectively. These systematic changes of alkene, alkadiene, and branched alkane contents were even more pronounced when the hydrocarbon fraction (fraction 1) alone was analysed. In this case, the contributions of the three substance classes to the total of hydrocarbons increased from 24 % ± 1.1, 0.38 % ± 0.071 and 0.05 % ± 0.048 in wax scales, to 40 % ± 0.5, 7.0 % ± 0.09 and 3.3 % ± 0.08 in old comb wax respectively (Table 16.7).

Table 16.7 Relative chemical composition of A. m. carnica hydrocarbon fractions of comb waxes of different ages (Frölich et al. 2000)

More recently, Namdar et al. (2007) published GC and GC/MS analyses of light and dark coloured A. m. ligustica and A. m. syriaca combs (Fig. 16.3). They found that, as beeswax ages and darkens, its n-alkane composition changes. The amount of even numbered n-alkanes (C₂₂–C₃₂), is significantly higher in darker coloured beeswax compared to light beeswax. They attributed these differences, at least in part, to the accumulation of cuticular residues known to contain C₂₃ to C₃₂ odd and even numbered n-alkanes. They determined the presence of odd and even numbered n-alkanes, and showed that there was a clear predominance of the C₂₇ alkane, with only very small amounts of even numbered n-alkanes in the range of C₂₂–C₃₂. Also, darker beeswax contains on average about 3 times more even numbered n-alkanes than lighter coloured beeswax.

Fig. 16.3

Histogram of the averaged peak areas of the alkanes extracted from light coloured (white columns) and dark coloured (black columns) A. m. ligustica beeswax samples. The relative peak areas are normalized to the most abundant alkane. Cx refers to n-alkane with x carbons in its chain. Y axis =  % (from Namdar et al. 2007)

16.3.2 Discrimination and Classification of Beeswaxes

Before introducing this topic, it is often important to identify and separate pure beeswax from contaminant resins, such as slumgum, which occur in beeswax samples (Grout 1946; Morales-Corts et al. 2010). It was recently reported that waxes and contaminating resins can readily be identified by differential scanning calorimetry (Zhang et al. 2012). Quantitative criteria for the distinction between comb age classes, castes are possible based on chemical features of the respective waxes are both desirable and possible Frölich et al. (2000) subjected their data to a discriminant function analysis which allows the predictive classification of cases (wax samples) by computation of classification functions. These functions are not to be confused with discriminant functions. Only substance classes that could be positively identified by gas chromatography-mass spectrometry, were included. The results of their analysis functions achieved 99.3 % unambiguous discrimination into the classes: wax scales, new wax, middle aged wax and old wax.

The chemical changes recorded by Frölich et al. (2000) during the ageing process of comb wax, seem to consist of two parallel processes. They proposed that the decrease in chain length with age (process 1), may be due to lipolytic enzymes (Kurstjens et al. 1985; Davidson and Hepburn 1986; Hepburn 1986), which bees add to the wax scales during their conversion into comb wax. These enzymes might be esterases, and this could result in a decrease in long-chain esters and subsequently an increase in shorter chains. The second process (2), may be due to spontaneous physical and chemical processes rather than the direct influence of the bees. The olfactory system of the honeybee is very sensitive to hydrocarbon compounds (Page et al. 1991), the clearly distinguishable wax compositions may be cues for the honeybees to distinguish different regions of the nest for allocating tasks, or to identify nestmate bees they meet in the darkness of the nest (Tautz 2009) (cf. Chap. 5). Phiancharoen et al. (2011) calculated the weighted frequency distributions of the compounds in Table 16.4 to determine the average chain length of each type of wax as shown in Table 16.8. There were no significant differences among the waxes, although there is a trend suggesting that the waxes of the dwarf honeybees have the longest chain lengths. This is surprising because, as a general rule, stiffness, strength, yield stress and other properties increase with increasing carbon chain length in polymers (Salamone 1996), but this relationship does not hold for beeswaxes.

Table 16.8 Weighted frequency distributions for carbon chain length variation in A. mellifera, A. cerana, A. florea, A. andreniformis, A. dorsata and A. laboriosa waxes calculated from the data in Table 16.4 (Phiancharoen et al. 2011)

In a further study on wax discrimination Phiancharoen et al. (2011) performed a cluster analysis of beeswax composition, based on the data of Aichholz and Lorbeer (1999) (Table 16.4) to assess their relative affinities, as measured by the Euclidean distances using the unweighted pair-group centroid amalgamation rule. A parsimonious unweighted pair-group analysis based on the distribution of the chemical constituents for 82 elution peaks of the derivatized comb waxes of A. andreniformis, A. cerana, A. dorsata, A. florea, A. laboriosa and A. mellifera is shown in Fig. 16.4. The giant honeybee group (A. dorsata and A. laboriosa) is clearly separated from the other species, as are the dwarf species (A. andreniformis and A. florea), while A. mellifera is placed close to its sister-group, A. cerana.

Fig. 16.4

Hierarchical clustering diagram derived from single linkage clustering compound composition of derivatized comb wax (Phiancharoen et al. 2011)

The Euclidean distances of beeswaxes presented a very similar picture, which is consistent with the recent analyses of Apis species, in which three distinct clusters of sister-groups result from morphometric (Alexander 1991), behavioural (Raffiudin and Crozier 2007) and DNA sequence analyses (Arias and Sheppard 2005): (1) dwarf bees (A. andreniformis and A. florea); (2) giant honeybees (A. dorsata and A. laboriosa); and (3) a cluster consisting of the medium-sized bees (A. cerana, A. koschevnikovi, A. mellifera, A. nigrocincta and A. nuluensis). In any event, the close proximity of the beeswax unweighted pair-groups to those based on DNA and morphometrics, suggests that the wax glands and the products of secretions were highly conserved features during honeybee evolution (Fig. 16.4).

16.4 The Proteins of Beeswax

That beeswax might contain non-lipoidal material has been a very real possibility since Huber (1814) showed that beeswax scales and comb wax have different solubility characteristics. A century later Lineburg (1924) described in detail how worker bees chew and maul wax scales, adding a frothy substance to them. Kurstjens et al. (1985) pursued this probability as a by-product of their studies on the physical changes that occur in the conversion of wax scales into fashioned comb. They found that scale wax did not exhibit a detectable monoglyceride fraction, but had a relatively large pool of diglycerides. In comb wax there was a pronounced monoglyceride fraction, and the diglyceride fraction was considerably less than that in scale wax.

These gross chemical differences between wax scales and finished combs led directly to a search for proteinaceous material that could be added to the wax during chewing, and which might have the expected lytic properties, as had been noted decades earlier by Lineburg (1924). In the search for bee-derived proteins in beeswax, it was essential to preclude any contamination of the scale and comb waxes used in the analyses. Such wax was obtained by keeping small colonies of bees made from newly enclosed brood, confined in a laboratory with no opportunity to forage, nor access to pollen or honey. The bees were only fed a syrupy solution of sucrose. Kurstjens et al. (1985) were able to confirm that scale wax obtained under these conditions contained about 2 μg of protein /mg of wax, and that comb wax contained about 6 μg of protein/mg of wax.

Because beeswax is hydrophobic, it was surmised that it is transported through the pore canals to the exterior surface of the wax mirror by lipophorins. This appears to be the major transport mechanism of hydrophobic natural products in insects (Gilbert and Chino 1974; Haruhito and Chino 1982). Because the lipid composition changes in the conversion of scales into comb wax (Kurstjens et al. 1985), it is also likely that some lipolytic protein is introduced into the scale wax when the bees chew it (Lineburg 1924; Kurstjens et al. 1985). In a series of electrophoretic studies on the beeswax proteins of A. m. capensis and A. m. scutellata, Kurstjens et al. (1990) showed that the substructures of the wax scale and comb protein fractions contained 11 and 13 bands respectively. Seven of these bands were common to both scale and comb waxes for both subspecies.

The proteins ranged between 19 and 100 kD. Bands 1, 2, 6 and 17 (about 97, 89, 66, and 19 kD respectively), were unique to scale wax, while bands 3, 4, 10, 11 and 15 (87, 82, 54, 47 and 43 kD respectively), were unique to comb wax. The waxes shared bands 5, 7–9, 12, 14 and 16 (70, 60, 57, 55, 51, 44 and 29 kD respectively). The densitometric scans showed the relative molecular weight distributions of the bands, and that band 17 is dominant in scale wax, while bands 7–12 are collectively dominant in comb wax. Although wax scales and comb wax contain both unique and shared proteins, their functions are unknown. However, two kinds of lipophorins occur in honeybees (Ryan et al. 1984), and it was surmised that apolipophorin II of honeybees at 78 kD is very close to the 82-kD fraction of comb wax, and to the 70-kD fractions shared by both comb and scale waxes. Although workers chew wax during comb-building, sometimes almost intact scales can be seen in cell walls (Casteel 1912; Zhang et al. 2010), this too points to the addition of a salivary secretion because when incorporated in scale wax, the diacylglycerol component of scales is reduced, and the monoacylglycerol fraction of comb wax increases (Davidson and Hepburn 1986).

16.5 Plant-Derived Aromatic Volatiles and Colourants in Beeswax

Although beeswax has long been a very valuable commodity and its aroma one of its particularly favoured qualities, no analyses of these volatiles were undertaken until the work of Ferber and Nursten (1977). They used a combined GC-MS approach, and for positive identification, they used retention indices of ±0.10 for unknowns and standards on each of two columns of differing polarity, as well as acceptable mass spectral data (Table 16.9). In view of the now well-established interactions between pheromones and comb and/or cuticular waxes (Breed et al. 1995a, b, 1998), it is essential to know the chemical composition of the waxes involved, and to be able to classify same. The aromatic volatiles detected in A. mellifera wax and listed by Ferber and Nursten (1977) could lead to unimagined possibilities for studies on nestmate recognition.

Table 16.9 Volatile components of beeswax characterized by gas chromatography-mass spectrometry (Ferber and Nursten 1977)

Subsequently Puleo (1991) performed an exhaustive analysis of the minor constituents of beeswax. Table 16.10 demonstrates the extraordinary diversity of plant-derived compounds (collectively, propolis). Among them is a large percentage of chromophoric (C = C, C = O, N = N, C–NO₂₎ and auxochromic (C–OH, CNH_2, COOH) groups, which contribute to the strong colour of beeswax. This results from the fact that the auxochromes enhance the colouring capacity of the chromophores (Puleo 1991).

Table 16.10 Components of propolis recovered from beeswax (Puleo 1991, and references therein)