Differences in Clutch-Size and Reproductive Behavior in Show Versus ...

3 downloads 0 Views 251KB Size Report
Differences in Clutch-Size and Reproductive Behavior in. Show Versus Pet Budgerigars (Melopsittacus undulates). SABINE G. GEBHARDT-HENRICH and ...
Bird Behavior, Vol. 17, pp. 00–00 Printed in the USA. All rights reserved

1056-1383/05 $20.00 + .00 Copyright © 2005 Cognizant Comm. Corp. www.cognizantcommunication.com

Differences in Clutch-Size and Reproductive Behavior in Show Versus Pet Budgerigars (Melopsittacus undulates) SABINE G. GEBHARDT-HENRICH and ANDREAS STEIGER Division of Animal Housing and Welfare, Institute of Animal Genetics, Nutrition and Housing, Vetsuisse Faculty of the University of Bern, CH-3001 Bern, Switzerland The aim of this study was to compare the reproductive behavior of large budgerigars bred by breeders attending exhibitions (show type) with smaller pet budgerigars (normal type). Compared with the normal type, show-type budgerigars were significantly heavier, had longer tarsi, longer wing feathers, longer feathers on the top of the head, and significantly larger spots on their throat feathers. Show-type females had a lower ratio of body mass to structural size (i.e., condition) than did normal-type females; such a low condition was associated with smaller clutches and egg cannibalism. Females paired with show-type males were less likely to lay eggs than those paired with normal-type males, although morphological traits within a pair were uncorrelated. Although the influence of male behavior on female reproductive behavior is unknown, all traits selected for in male show birds were negatively correlated with the frequency of flying from perch to perch and with reproductive parameters. Thus, there were some indications that budgerigars close to the breeders’ ideal were less successful in reproducing than budgerigars resembling the wild type. Key words: Behavior; Budgerigar; Exaggerated character; Reproduction; Captive breeding

Artificial selection for exaggerated characters in domestic animals has been criticized as being harmful in various animal groups (Bartels & Wegner, 1998; McGreevy & Nicholas, 1999; Not Schläpfer, 1998; Peyer, 1997; Stucki, 1998 and references therein). For example, breeders have selected for changes in the appearance of the budgerigar (Melopsittacus undulates) for the purpose of exhibiting these birds at competitions in many countries (Vins, 1993). The British standard of the exhibition budgerigar (Fig. 1) calls for a large bird with long feathers (http://www.worldbudgerigar.org/). Breeders’ standard is for the head to appear large and, when observed from the front, for feathers to cover the

eyes. The throat feathers (part of the so-called mask) should be long with large distinct spots. Several authors have mentioned problems regarding animal welfare in the captive breeding of budgerigars. For example, various feather diseases seem to be prevalent in British exhibition budgerigars (Baker, 1996a), which also have a shorter life expectancy than budgerigars bred for the general pet trade (Baker, 1996b, 1997). Although there are no data, veterinarians and breeders claim that exhibition budgerigars also are more prone to diseases and are less fertile than the wild type (Gerlach, 1994; Vins, 1993). Our aim in this study was to systematically compare the reproductive biology, including

1

2

GEBHARDT-HENRICH AND STEIGER

wild type

show type

Figure 1. Show-type (right) and wild-type (left) budgerigars. The standard of the exhibition budgerigar (show type) calls for a large bird with long feathers almost covering the eyes and large spots on the throat (not drawn to scale).

length of the third primary wing feather, length of several top head feathers, and vertical length of the largest spot at the throat. Tarsus length is considered the best measure of structural size in birds (Rising & Somers, 1989). Feather length is a selection criterion in exhibition budgerigars. The length of the primary wing feathers might influence flight performance, and both head feathers and throat spots are believed to be important during courtship. Birds were weighed with an electronic balance to the nearest 0.01 g; tarsus length and the length of the spot were measured by using calipers. The length of the wing feather was measured with a ruler, and the length of the head feathers was measured with a cardboard grid that was moved forward so the feathers were raised. The index of condition was calculated as body mass divided by the cube of tarsus length. For scaling purposes, this index was multiplied by 1000. Similarly, the index of wing loading was calculated as body mass divided by the cube of the length of the third primary feather. For scaling this index was multiplied by 1,000,000. Selection and Pairing of Birds

behavior, in budgerigars bred for exhibitions (show type) and budgerigars bred for the pet trade (normal type) and to test whether reproduction was poorer in show types. Subjects and Methods Birds and Husbandry Budgerigars were purchased in Northern Switzerland. Some were obtained from breeders who exhibited their birds (show birds) and others from breeders whose birds were not bred for exhibition (normal birds). After purchase, the budgerigars were initially kept as one flock in a large 5 × 2 × 2-m aviary for more than 4 months. They were fed daily a commercial budgerigar food and germinated budgerigar food; vitamins were added to the drinking water once a week. Grit and mineral supplements were available ad libitum. Lights were on between 0700 and 1615 h CET with a dusk phase. Additionally, natural light illuminated the room. Food was available in several dishes on the litter-covered aviary floor, and water was available in rodent drinking bottles. Shortly before transferring the birds to breeding boxes, measurements were made of body mass, tarsometatarsus (hereafter referred to as tarsus) length,

Truly random pairing might have resulted in mates of such different sizes that copulation might have been impossible. In addition to the breeders’ classification into show birds and normal birds we did a cluster analysis using our measurements of the birds. Approximate covariance estimation for clustering analyses [PROC ACECLUS (SAS Inc., 1988)] was performed for males and females separately, and after visual inspection of the plot of the canonical coefficient 2 on the canonical coefficient 1, birds were classified into either s-type (heavy, long feathers) or n-type (smaller body mass, shorter feathers). In males 88% of the variance was explained by the first two eigenvectors, and two distinct clusters emerged (Fig. 2). In females, two clusters emerged only after we omitted body mass and head feather length from the analysis (Fig. 3). Within the s- and n-types, birds were paired randomly, but identical colors within one pair were avoided in order to visually distinguish the sexes on video. For the statistical analyses, we used the original breeder classification. Breeding Twenty-eight pairs of budgerigars were placed pair-wise in commercial boxes (cabinets) of

canonical coefficient 2

BUDGERIGAR BEHAVIOR

3

9

4

0

-4

-8 -20

-10

0

10

20

canonical coefficient 1

canonical coefficient 2

Figure 2. The canonical coefficient 2 is plotted against the canonical coefficient 1 in males. The canonical coefficients were obtained by the approximate covariance estimation for clustering analyses (see text). All purchased birds are shown; therefore, the number on the plots exceeds the sample sizes of the study. N designates birds obtained by breeders not attending exhibitions, S designates birds obtained by breeders who exhibit their birds.

20

10

0

-10

-20 -20

-10

0

10

canonical coefficient 1 Figure 3. The canonical coefficient 2 is plotted against the canonical coefficient 1 in females.

20

30

4

GEBHARDT-HENRICH AND STEIGER

80 × 40 × 50 cm, with wire at the front and litter on the floor. Two perches were spaced 51 cm apart. Husbandry conditions were the same as mentioned above. The pairs could not see other birds, but could hear other budgerigars that were kept in the same room. The birds were put into boxes in early September 2000 and all filming was done in early October. Feathers covering the cloaca of females were cut to ensure the possibility of fertilization. This is standard practice when breeding show budgerigars (Bartels, personal communication). We checked nestboxes daily for eggs, which we weighed. Eggs were candled after 1 week and unfertilized eggs were removed. Nestlings were weighed every day until fledging. Observations After about 2 weeks, wooden nestboxes (27 × 17 × 15 cm) were attached to the breeding boxes and videotaping began 3 days later. Because budgerigars are more active in the morning (personal observation, and Stamps, Clark, & Arrowood, 1985), we videotaped the pairs for 30 min between 0900 and 1200 h. Four pairs were filmed simultaneously with four cameras, and some pairs were filmed on two different days for repeatability checks. During filming, breeding boxes were additionally illuminated by nonflickering neon lights. Analysis of Behavior Videotapes were analyzed using Observer® Video-Pro Version 4.0 (Noldus Information Technology) software. The names of the behavioral elements follow Brockway (1963, 1964). We measured the frequency and duration of “states,” which included eating, preening (including foot preening), reciprocal preening, and courtship feeding. The frequency of “events” was measured, which include flying, copulation, climbing at the wire, bill thrust, nestbox inspection, and head bobbing. Head bobbing was scored as either directed towards or away from the mate. Courtship behavior was the sum of the frequencies of head bobbing towards the mate, reciprocal preening, and courtship feeding. Only one recording of each pair was used in the analyses. Second recordings of the same pair were only used in the repeatability analyses (frequencies of behaviors were correlated among the recordings). Parametric tests were performed using PROC GLM

of the SAS software package, and nonparametric tests were performed with PROC NPAR1WAY (SAS Inc., 1988). Data on the frequency of flying were log10-transformed. We measured flying speed in 160-cm-long boxes. The birds were videotaped, and frame-by-frame analyses determined the exact time when the bird left one perch and when it reached the opposite perch (Observer Software by Noldus). Results Morphological Differences Among the Types Both male and female budgerigars purchased from breeders (show type) had significantly greater body mass, longer tarsi, longer wing feathers (third primary feather), larger black spots on their throat feathers, and longer head feathers than normal-type budgerigars (Fig. 1, Table 1). In females, show-type females had a significantly lower index of condition than normal birds, but wing loading did not differ between show and normal types (Table 1). This means that show-type females weighed less relative to their structural body size than did normal-type females. In males, normal and show types neither differed in condition nor wing loading (Table 1). The flying speed of males and females was negatively correlated with condition and wing loading (loading: p = 0.0039, r2 = 0.64; condition: p = 0.0007, r2 = 0.62), but did not differ between show and normal types. Cluster analysis did not change the classification of males with two exceptions (i.e., one show-type male was classified as n-type and one male from the normal type was classified as s-type), and the new classification did not change the morphological differences between the groups. As mentioned before, females could not be grouped into two clusters when all body measurements were considered. In the new classification based on the cluster analysis, five “normal” females were now classified as s-type and three show females were classified as n-type. This new classification of the females did not match the visual appearance of the birds. Body mass was correlated with tarsus length in both sexes (Table 2). In females tarsus length, wing feather length, head feather length and spot size were positively correlated (Table 2). In males correlations were similar, except tarsus length was not correlated with the wing and head feather length and spot size

BUDGERIGAR BEHAVIOR

5

Table 1. Measurements of Female and Male Budgerigars of the Normal and the Show Type Before Mating Trait Females Body mass (g) Tarsus (mm) Wing (mm) Spot (mm) Head (mm) Condition Wing loading Males Body mass (g) Tarsus (mm) Wing (mm) Spot (mm) Head (mm) Condition Wing loading

Show type (N)

Normal type (N)

Statistic

p-Value

62.90 (12) 14.97 (12) 79.48 (12) 5.8 (10) 11.38 (12) 18.84 (12) 126.44 (12)

56.34 (16) 13.96 (16) 74.45 (15) 3.8 (16) 9.19 (16) 20.64 (16) 137.20 (15)

F-value: 4.63 F-value: 15.13 F-value: 15.48 Wilcoxon F-value: 23.70 F-value: 4.47 Wilcoxon

0.04 0.0006 0.0006 0.0025 0.0001 0.04 ns

60.54 (14) 14.89 (14) 82.5 (14) 5.2 (14) 12.86 (14) 18.43 (14) 108.93 (14)

50.53 14.09 77.42 3.88 10.32 18.12 109.08

F-value: 12.12 F-value: 14.45 F-value: 13.57 F-value: 8.39 Wilcoxon F-value: 0.08 Wilcoxon

0.0018 0.0008 0.001 0.0079 0.001 ns ns

(14) (14) (14) (12) (14) (14) (14)

Note: This table is based on the classification of the breeders from whom the birds were purchased. Condition (g/mm3 × 1000) was calculated by dividing body mass by tarsus length3 and wing loading (g/mm3 × 1,000,000) was calculated by dividing body mass by the length of the third primary wing feather3. N = number of measured females and males.

(Table 2). For n- and s-type birds combined, traits were not correlated between the mates except for the length of the wing feather (r = 0.48, n = 27, p = 0.01). Within n- and s-types, the length of the wing feather was not correlated among the mates. Reproduction Within 6 weeks after the introduction of nestboxes, only 75% of all pairs had started a clutch. Consistent with the pet literature (Gerlach, 1994; Vins, 1993), females paired with normal males were more likely to have a clutch than females paired with show males (Table 3). Males paired with females with a clutch had a significantly shorter tarsus than males whose females had no clutch [14.31 vs. 15.01 mm, F(1, 26) = 6.56, p = 0.017]. On the other hand, females of either type were equally likely to start a clutch. Females with versus without a clutch did not differ in any of the traits measured in this study. However, among females with at least one egg, clutch size was positively correlated with their condition measured before pairing (r2 = 0.33, N = 20, p = 0.0067) (Fig. 4). Three females with very low conditions started a clutch, but later destroyed or ate their eggs.

Morphological differences between the mates within a pair did not influence the probability of a clutch or the fertilization rate of the eggs, with one exception. Pairs that had larger differences in spot size had fewer clutches (absolute difference of spot size in pairs with a clutch: 1.12 ± 0.95 cm, n = 18; without a clutch: 2.83 ± 0.86 cm, n = 6, Wilcoxon test p = 0.006). Spot size in males was significantly positively correlated with the difference in spot size between members of the pair (Spearman rs = 0.53, p = 0.008, n = 24). Males without clutches had significantly larger throat spots than did males with

Table 2. Correlation Among Different Body Traits Males Females Body mass Tarsus Wing Spot Head

Body Mass

*** ns ns ns

Tarsus

Wing

Spot

Head

*

** ns

* ns *

*** ns * ns

* * **

** **

*

Significance levels: *p < 0.05, **p < 0.01, ***p < 0.001.

6

GEBHARDT-HENRICH AND STEIGER

Table 3. Relationship of Females Paired With NormalType Versus Show-Type Males and Incidence of Clutches Type of Male Normal Show Total

No Clutch

Clutch

Total

1 6 7

13 8 21

14 14 28

Note: Fisher’s exact test (1 df), one-sided p = 0.038.

clutches [5.66 vs. 4.19 mm, F(1, 24) = 7.99, p = 0.009]. Females with and without clutches did not differ in throat spot size nor was spot size in females correlated with the difference in spot size within a pair. Because males without clutches had longer tarsi, as well as larger spots and predominantly belonged to the show type, it is impossible to separate the effects of tarsus, spot, and type statistically. Clutch size and egg masses were not correlated. Egg mass, however, was positively correlated with female body mass and tarsus length (body mass: rs = 0.63, n = 21, p = 0.002; tarsus: rs = 0.51, n = 21,

p = 0.02). Egg mass was not correlated with the condition of the female (p = 0.86). Show-type females had heavier eggs (show type: 2.70 g, normal type: 2.50 g) (Table 4). Fertilized and unfertilized eggs did not differ in mass. Hatching mass correlated with egg mass and differed among pairs, but not among types of females [type: F(1, 20) = 2.27, p = 0.15; female (each female belonged to either type): F(7, 20) = 3.14, p = 0.02; egg mass: F(1, 21) = 36.32, p < 0.0001]. Almost all nestlings died before fledging. The likely cause was an infection with the avian polyoma virus and the circovirus. The avian circovirus is the etiological agent of the psittacine beak and feather disease (PBFD). The presence of the avian polyoma virus was confirmed by PCR. The age at death of the nestlings was not correlated with any measured traits of the parent birds. Over 3 years, adult mortality in our stock did not differ between the types when age (year of the band) was controlled (Cochran-Mantel-Haenzel-Statistics = 0.224, 1 df, ns). The exact age was unknown, but the year stamped on the band was a good indication of age, because birds were banded only as nestlings.

6

5

eggs

4

3 2 1 14

15

16

17

18

19

20

21

22

23

24

Condition [g/mm3] x 1000 Figure 4. Relationship between number of eggs and condition [body mass (g)/tarsus length (mm3) × 1000] before mating. Filled circles indicate females that ate their eggs. Open circles and dashed line: show-type females. Asterisks and solid line: normal-type females. The difference between the types as well as the interaction between type and condition were not significant.

BUDGERIGAR BEHAVIOR Table 4. Differences in Egg Masses (g) Among Females and Among Types

Type Female (type) Error Total

df

Mean Square

F-Value

p-Value

1 7 21 29

0.3111 0.3194 0.0224

13.87 14.23

0.0013 0.2) and was therefore pooled. When analyzing only males that had a future clutch, the difference in flying from perch to perch between the two types remained nonsignificant [F(1, 19) = 3.73, p = 0.069]. The trend of males of the show type to fly less often was reflected by the negative correlations between different body size traits and the frequency of flying. The closer the male budgerigar was to the ideal show bird (i.e., large with long feathers and spots), the less frequent was flying (Table 6). Condition and wing loading was not associated with flying frequency (Table 6). However, because more show-type males had no future clutches, the effect of type or reproductive state on the frequency of flying could not be separated. Birds of both sexes with future clutches inspected the nestboxes more often than did birds without future clutches (males: Wilcoxon z = –2.2121, p = 0.035; females: Wilcoxon z = –2.2575, p = 0.0323). Males with and without future clutches exhibited comparable levels of courtship behavior. There was no significant difference in the frequency of courtship between the different types of males. There was, however, a significant negative correlation between

8

GEBHARDT-HENRICH AND STEIGER

Table 6. Spearman Rank Correlations Between Traits of the Males and the Frequencies of Flying From Perch to Perch Trait

Flying From Perch to Perch

Body mass (g) Tarsus (mm) Wing feather (mm) Spot (mm) Head feather (mm) Condition (mm/g3 × 1,000,000) Wing loading (mm/g3 × 1,000,000)

–0.451 –0.336 –0.340 –0.397 –0.327 –0.202 –0.036

(0.016) (0.080) (0.077) (0.045) (0.089) (ns) (NS)

Note: Tarsus length, wing feather, spot, and head feather were measured in millimeters. Condition was calculated as body mass/ tarsus3; wing loading was calculated as body mass/wing feather3. N = 28. p-Values are shown in parentheses; ns: p > 0.2.

throat spot size and the number of courtship acts (rs = –0.423, p = 0.0315, n = 26). Males having larger throat spots courted less. The frequency of head bobbing toward the females and male courtship behavior (including head bobbing) was positively correlated with the fertility rate of the future eggs (courtship: rs = 0.694, p = 0.0007; head bobbing: rs = 0.5535, p = 0.0113, n = 20); thus, the more the male courted his mate, the more future eggs were fertilized. Even when excluding one male that exhibited extremely frequent head bobbing, the correlations did not change significantly (courtship: rs = 0.671, p = 0.0017; head bobbing: rs = 0.506, p = 0.027, n = 19). There were no correlations between the number of courtship acts and morphological traits of the female or morphological differences within the pair. Discussion Budgerigars have been bred in captivity since the end of the 18th century. Wild budgerigars are much smaller than domesticated budgerigars. In a field study, Wyndham (1980) found that both sexes weighed less than 30 g, the males being significantly heavier than females. The lean dry mass without gonads did not differ among the sexes, contrasting with our finding of larger body mass of females in our study. One reason for the greater body mass of females could be their breeding condition. According to Wyndham, Hutchison, and Brockway-Fuller (1981), adult domesticated budgerigars always have

developed gonads, whereas in wild budgerigars gonads regress outside the breeding season. However, we had no healthy budgerigar of either sex weighing as little or close to the mean body mass reported for wild budgerigars by Wyndham (1980). The mean body mass of our budgerigars was about twice as much as the body mass of wild-caught birds. Small “normal” birds and big show birds were not discrete types. All traits selected for by show breeders, such as large size, long (head) feathers, and large spots on the throat, were distributed continuously among our birds. One possible reason underlying this continuous distribution is that such traits are quantitative and influenced by many genes and continuously varying environmental factors. Also, there is no reproductive isolation between the types. Surplus show birds are often sold to pet stores and interbreed with “normal” birds. In our study, we were unable to obtain the extreme forms because exhibition breeders were not willing to sell their most successful birds. It is possible that budgerigars bred for exhibition or for the pet trade differ in other aspects not reflected by their morphological differences. Nevertheless, our birds were highly variable in the traits that are selected in show types. Even without the extreme types, we found significant correlations of the morphological traits with reproductive parameters. Our study suggests possible reproductive impairments in show budgerigars Show-type females had a lower condition at the beginning of breeding and laid fewer eggs. The three females that ate their few eggs had a low condition and were all show birds. Because show breeders favor large birds, such large females might have difficulty accumulating sufficient nutrients under our aviary conditions. Considering that large females were more than twice as heavy as wild females, selection for body mass may not have been reflected in the ability to digest more nutrients. Larger females also laid larger eggs, so the uptake of calcium might have been a limiting factor. More show-type males were paired with females without clutches. Because mating was within s- and n-types and the behavior of the mates was strongly correlated, it is not possible to attribute the presence or absence of a clutch to the type of male rather than the female. Show-type females might have differed in ways that were not obvious to us. Furthermore, the birds most resembling the successful show types

BUDGERIGAR BEHAVIOR were largely males. However, except for wing length, the traits between two mates within a pair were uncorrelated. This suggests that the show-type males reproduced less often than normal males. In budgerigars, females choose mates (Brockway, 1964; Cinat-Tomson, 1926). Female budgerigars forced to mate with (unattractive) males lay smaller clutches or no clutch at all, and the hatchability of the eggs is lower (Massa, Galanti, & Bottoni, 1996). Both Engesser (1977) and Cinat-Tomson (1926) found that female budgerigars were less likely to choose passive and calm males. In our birds, flying from perch to perch appeared to be linked to courtship behavior. Males flew more often than females, and those males were more likely to have a clutch later on. It is not clear if the low frequency of flying had a direct influence on the female’s reproductive behavior, or what may have caused little flying. However, all selected traits in show birds were negatively correlated with the frequency of flying and reproduction. Interestingly, Cinat-Tomson (1926) found that females preferred active males with many spots. Spot number and activity level were positively correlated (Cinat-Tomson, 1926). After manipulation of spot number, less active males were less preferred even with added spots (Cinat-Tomson 1926). The birds in that study were most likely not show type, because the show type was developed after 1945 in England (Binggeli, 2002). Modern selection in showtype males appears to have resulted in a negative rather than positive correlation between spots and activity. Our show-type birds had larger and more numerous spots (personal observation). There are various reasons underlying the observed differences between show-type and normal-type budgerigars. Because our birds were not bred under identical conditions, systematic differences in raising and husbandry among breeders might have led to the observed differences. However, it is not reasonable to assume that show breeders of valuable birds would keep them in poorer condition than would other breeders. All birds had been kept in a large aviary for about 3 months between purchase and the start of this study. Probably, differences in the strength and the kind of selection between the two populations accounted for at least part of the difference in breeding performance. Whereas in the pet population breeders selected birds that had more offspring, breeders of show birds selected them ac-

9

cording to certain desired traits even if these birds were poor breeders. Thus, in the show population “natural” selection on breeding performance was much weaker. These differences in selection pressures between the two populations could account for the observed results (e.g., the degeneration of breeding-related traits in the show population). Additionally or alternatively, the selected traits could be associated with poor breeding. Breeders attending exhibitions might directly select for inactive birds, because they can be exhibited more easily. This possibility can only be tested by manipulating one trait at a time and keeping other traits constant. In conclusion, this study suggests possible negative influences on egg-laying performance in female show-type budgerigars, as well as in females paired to male show-type budgerigars. The lower condition of show-type females might lead to smaller clutches. Females paired with show-type males were less likely to lay eggs. Whether or not decreased flying activity of the show males is causally linked to lack of egg laying remains to be tested. The reduced incidence of reproductive behavior in show-type birds suggests that certain selection criteria can be detrimental. Author Note Our research on companion animals was made possible by the Swiss Federal Veterinary Office. Comments of Thomas Bartels, Petra Keller, anonymous reviewers, and the editor greatly improved the manuscript. Correspondence concerning this article should be sent to Sabine Gebhardt-Henrich, Institute of Animal Genetics, Nutrition and Housing, Division of Animal Housing and Welfare, P.O. Box, CH-3001 Bern, Switzerland. Electronic mail may be sent via Internet to [email protected] References Baker, J. R. (1996a). Survey of feather diseases of exhibition budgerigars in the United Kingdom. The Veterinary Record, 139, 590–594. Baker, J. R. (1996b). Causes of mortality and morbidity in exhibition budgerigars in the United Kingdom. The Veterinary Record, 139, 156–162. Baker, J. R. (1997). Mortality and morbidity in exhibition budgerigars (Melopsittacus undulates) in the U.K. In The 4th Conference of the European Committee of the Association of Avian Veterinarians (pp. 115–118). Bartels, T., & Wegner, W. (1998). Fehlentwicklungen in der

10

GEBHARDT-HENRICH AND STEIGER

Haustierzucht. Stuttgart, Germany: Ferdinand Enke Verlag. Binggeli, T. (2002). Gedanken zur Schauwellensittichzucht. Tierwelt, No. 31, 17. Brockway, B. F. (1963). Ethological studies of the budgerigar (Melopsittacus undulates): Non-reproductive behaviour. Behaviour, 22, 193–222. Brockway, B. F. (1964). Ethological studies of the Budgerigar: Reproductive behaviour. Behaviour, 23, 295–324. Cinat-Tomson, H. (1926). Die geschlechtliche Zuchtwahl beim Wellensittich (Melopsittacus undulates Shaw). Biologisches Zentralblatt, 46, 543–552. Engesser, U. (1977). Sozialisation junger Wellensittiche (Melopsittacus undulates Shaw). Zeitschrift für Tierpsychologie, 43, 68–105. Gerlach, H. (1994). Zuchtbedingte Anomalien bei Ziervögeln. Tierßrztliche Praxis, 22, 319–323. Massa, R., Galanti, V., & Bottoni, L. (1996). Mate choice and reproductive success in the domesticated budgerigar, Melopsittacus undulates. Italian Journal of Zoology, 63, 243–246. McGreevy, P. D., & Nicholas, F. W. (1999). Some practical solutions to welfare problems in dog breeding. Animal Welfare, 8, 329–341. Noldus Information Technology. (1997). The Observer, support package for video analysis, version 4.0 for Windows edition. Wageningen: The Netherlands.

Not Schläpfer, I. (1998). Beurteilung verschiedener Zuchtlinien von Ziervögeln, Kleinnagern, Zierfischen und Reptilien in tierschützerischer Hinsicht. Dissertation, Veterinär-medizinische Fakultät, Universität Zürich. Peyer, N. (1997). Die Beurteilung zuchtbedingter Defekte bei Rassehunden in tierschützerischer Hinsicht. Dissertation, Veterinär-medizinische Fakultät, Universität Bern. Rising, J. D., & Somers, K. M. (1989). The measurement of overall body size in birds. Auk, 106, 666–674. SAS Inc. (1988). SAS/STAT* user’s guide, release 6.03 edition. Cary, NC: SAS Institute Inc. Stamps, J., Clark, A., Arrowood, P., & Kus, B. (1985). Parent–offspring conflict in budgerigars. Behaviour, 94, 1– 40. Stucki, F. (1998). Die Beurteilung zuchtbedingter Defekte bei Rassegeflügel, Rassetauben, Rassekaninchen und Rassekatzen in tierschützerischer Hinsicht. Veterinärmedizinische Fakultät, Universität Bern. Vins, T. (1993). Das Wellensittichbuch. Alfeld, Leine, Germany: Schaper. Wyndham, E. (1980). Diurnal cycle, behaviour and social organisation of the budgerigar Melopsittacus undulates. Emu, 80, 25–33. Wyndham, E., Hutchison, R. E., & Brockway-Fuller, B. F. (1981). Gonadal condition of non-breeding wild and domesticated budgerigars Melopsittacus undulates. Ibis, 123, 511–518.