Interrelationships between egg dimensions, pore numbers, incubation time, and adult body mass in Procellariiformes with special reference to the Antarctic Petrel Thalassoica antarctica * F. MEHLUM, H. RAHN, C. BECH A N D S. HAFTORN Mehlum, F., Rahn, H., Bech, C. & Haftom, S. 1986: Interrelationships between egg dimensions, pore numbers, incubation time, and adult body mass in Procellariiformes with special reference to the Antarctic Petrel Tbalassoica antarctica. Polar Research 511,s.. 53-58. Dimensions of eggs and eggshells and pore density are reported for eggs of the Antarctic Petrel Thalassoica antarctica(as well as eggs of the Snow Petrel Pagodroma nivea). Allometric relationships are used to estimate the initial egg mass, total number of pores and incubation time. The initial egg mass for Antarctic Petrel was estimated to 83 g. and the incubation time to 55 days. The total pore number in Antarctic Petrel was estimated to 4212 and in Snow Petrel 2654 pores per egg. Using data from 8 species of Procellariiformes, the relationship between total pore counts, initial egg mass and incubation time is described. Fridrjof Mehlum. Norsk Palarinstitutt. P.O. Bor 158, N-1330 Oslo Licfrhavn. N o w a y ; Hermann Rahn. Department of Phwio1og.v. State University ($ New York. BuJJalo. N . Y . 14214. U S A ; Claus Bech. Department of Zoology. University of Trondheim. N- 7055 Dragvoll. N o w a y ; Svein Haftom, The Museum. University of Trondheim. Erling Skakkesgate 47A. N-7000 Trondheim. Norway: October 1986 (revised February 1987). The initial egg mass at the time of laying is a n important reference point because of its correla- tion with various egg parameters, e.g. eggshell dimensions, number of pores in the eggshell, incubation time, and adult body mass. Allo- metric relationships that have been established allow one to predict many of these, and other functions, once the initial egg mass is established. Initial egg mass and various egg dimensions of the Antarctic Petrel have only scarcely been reported. Schonwetter (1 960) presents data based on a single egg only, and Orton (1968) gives values of mass, length and width, but d o not list the nuinber of specimens nor provide data for egg- shell dimensions. The data presented here were collected at the end of the incubation period and d o not represent initial egg mass. Eggs o f the order Procellariiformes lose water during their incubation period (Grant et al. 1982a), which is similar in amount to that of other birds (n = 8 I ) , and average 15% of the initial egg mass (Ar & Rahn 1980). In this study we report dimensions for 2 1 eggs and eggshells and pore density for 9 eggs of the Antarctic Petrel (as well as 16 and 3 eggs for dimensions of eggs and eggshells, respectively, of the Snow Petrel Pagodroma nivea), which allow one to estimate the initial egg mass based on allometric relationships of egg dimensions versus egg mass established for 69 species of Procellarii- formes. Once the initial egg mass is known, one can calculate the total number of pores per egg and, furthermore, predict from other allometric relationships the incubation time and adult body mass for this species. The eggs were collected during the Norwegian Antarctic Research Ex- pedition 1984185 at the breeding colony Svart- hamaren (71"53'S, 5" IO'E) located in Dronning Maud Land (Mehlum et al. 1985). Methods Length and width dimensions were obtained with calipers and read to the nearest 0.1 mm. * Publication No. 87of the Norwegian Antarctic Research Expediiions ( I 9841851. 54 Shell thickness was measured in each egg at twenty different points, using a ball-point caliper which could be read to 0.01 mm. The shell thick- ness measurement included the dried membranes adhering to the shell. Pore density was etablished as follows: Pieces of eggshell from various regions were boiled in 5% (g/l) KOH solutions for several minutes to remove all proteinaceous material. They were then etched by being dipped briefly in concentra- ted nitrid acid for a predetermined period of about 15 seconds, which enlarges the pores. After drying, the inside of the shell was painted with a concentrated water solution of Evans Blue dye. By capillary action this solution fills the pores which now become visible at the outer shell sur- face as blue spots against a white background and can now be counted with a dissecting scope. The counting iield was 0.25 cmz, and twenty fields representing various regions of each egg were counted (see Fig. I ) . These averaged values where multiplied by 4, expressing the pore density per cm?. The latter value is factored by the total surface area of the egg derived from the relation- ship A (cm2) = 4.835 W’.“’ (Paganelli et al. 1984) where W = initial egg mass (g). Fig. I . Photograph of a 0.25 cm’ shell area showing tjpical distribution of dye-stained pores which had Results and discussion Prediction of initial egg mass Table I shows egg mass, egg and shell dimensi- ons previously reported, and those obtained in this study. We have estimated the initial egg mass from the regression of length and width dimensi- ons as a function of egg mass based on 88 corre- lates (69 species) listed by Schonwetter (1960) for the order Procellariiformes shown in Fig. 2. The XSEE is 1.03 and 1.02 for the length and width correlates, respectively, where XSEE is the anti- log of SEE (standard error of estimate) = value by which Y is multiplied or divided. The projection of 70 mm length and 48 mm width (from Table 1 ) on Fig. 2 predicts a n initial egg mass of 80 and 86 g, respectively. Thus, a n average value of 83 g is a reasonable prediction for the initial egg mass from this particular nest site. The pore density and number ofpores The average pore density for nine eggs was 11.7/0.25 cm’, or 46.8 cm-’ (Table 2). The total egg surface calculated (see Methods) on the basis of an egg mass of 83 g = 90 cm2. Thus, the total number of pores = (46.8 x 90) or 42 12 pores per egg. Table 3 gives egg and eggshell dimensions for the Snow Petrel and a pore density of (10.7/0.25 cm2) or 42.8 C m 2 for three eggs. The surface area for the 47 g egg is 62 cm2 and the total pore number = (42.8 x 62) or 2654 pores per egg. I n Table 4 are listed the egg mass, incubation time, and pore number of nine species of the order Procellariiformes, including the two reported from this study. They have been ranked accor- ding t o egg mass. The number of pores is only 50-60% of that predicted from the regression established for 161 species (Ar & Rahn 1985). Thus, for a given egg mass the pores of the order Procellariiformes are significantly fewer, which appears to be related to their prolonged incubati- on time (see below). When pore numbers in Table 4 are regressed against the relative growth rate of embryos, expressed as egg mass divided by incubation time (g.day-’), the slope is essenti- ally 1.0, a relationship that describes the pore prevrouslv been enlarged by acid etching. number for birds in general (Ar & Rahn 1985). 5 5 Table I . Egg mass, egg and shell dimensions of the Antarctic Petrel Thalassoica antarctica. Egg Shell Reference Mass Length Width n Mass Thickn. n mm mm g mm Area g ~ Antarctic Continent 95 71.4 50.0 1 7.14 0.35 1 Schonwetter 1960 70.0 48.0 - - - - Orton 1968 Wilkes Station 90 Dronning Maud Land - 70. I 48.6 21 1.29 0.406 9 Thisstudy S.D (3.1) (1.5) (1.01) (0.032) 100 E E 0.361 Length 14.08 W n = 88 r = .998 10 10 5 50 r = ,998 SEE = X 1.02 I) 5 10 Fig. 2 . Regression of egg length and egg width upon egg mass f o r 69 species and 19 subspecies oj’the order Procellari$onnes takenfrom the tables of Schonwetter ( 1 960). Setting aside the unusally high pore count of Pterodroma hypoleuca in Table 4, the pore count of Procellariiformes is best described by the relationship, the number of pores, N = 3100 (W/I), S.D. 440, where W = initial egg mass, g, a n d I = incubation time, days. Incubation rime In Fig. 3 are plotted the incubation times against egg mass for 475 species excluding the order Procellaniformes (Grant et al. 1982a). The dotted lines mark the limits for k SEE. The shaded 56 Table 2 . Pore density (number of pores per field of 0.25 cm') for nine eggs o f the Antarctic Petrel Thalassoica anrarcrica. For each egg 20 fields were counted and averaged. Egg No. 2 3 4 5 7 8 9 10 II Grand Mean Pore density 10.1 12.0 13.3 15.3 9.8 10.4 11.7 13.5 8.8 11.7 S.D. 2.0 2.2 3.8 2.8 2.1 2.6 2.7 3.4 2.2 2. I Table 3 Egg mass. egg and shell dimensions, and pore density o f t h e Snow Petrel Pugodroma nivea Area M a s s Length Width n Mass Thickn. n Reference g mrn mrn g mm S. Georgia. S. Orkney 47.0 56.5 40.2 15 3.40 0.27 I5 Schonwetter 1960 Davis Station 47.4 55.5 39.4 21 - Dronning Maud Land - 56.7 39.2 16 3.42 0.28 3 Thisstudy - - Brown 1966 S.D ( 2 . 1 ) (1.0) Pore density (number of pores per 0.25 cm') Egg No. I 4 7 G r a n d m e a n Pore density 11.3 11.7 9.2 10.7 S.D. 2.8 2.8 2.2 1.3 Table 4. Egg mass, incubation time. and number of pores per egg in 9 species of the order Procellarii- formes. Species Egg Incu- N o . Refe- mass bation pores rence g days Diomedea exu1an.r )) nigripec u rrnmurabilrs Fulniarus glaciali.~ i7iala w ) i c a anrarcrica frerodroma phaeopygia Fullinus pac$nrs Pagodroma nivea Pterodroma hypoleuca 500 78 30.5 66 2 8 5 65 I 0 1 49 x 3 ( 5 5 ) 77 5 5 57 52 47 45 39 49 19.698 16.700 15.753 5.210 4.2 I 2 4.544 3.747 2.654 4.159 I 1 1 & - 3 4 5 6 4 7 ~ Re/erence.t : ( 1 ) Tullett & Board 1977; ( 2 ) Grant et al. I982a; ( 3 ) Rahn et al. 1984a: (4) this study; ( 5 ) Whittow et al. 1983:(6)Whittow 1 9 8 4 ; ( 7 ) G r a n t e t a l . 1982b. region above shows the regression .+ SEE for 38 species of the Procellariiformes where the incu- bation times were taken from Watson ( 1 975) and other sources, and the corresponding egg weight from Schonwetter (1960). For the Procellariifor- rnes the regression is-incubation time, I = 31 W0.l3'', + = 0.82 and X S E E = 1.1 1. For an 83 g egg of the Antarctic Petrel one would predict an incubation time of 55 days (range 50-61 days for f 1 SEE). This predicted value is larger than the reported incubation period of 40 -45 days (Orton 1968; J.A. van Franeker pers.comm.). The shortening of the incubation period compared to other Procellariiformes may be an adaptation for breeding in the hostile climate on the Antarctic continent. Adult body weight and egg mass From the egg weight (Schonwetter 1960) and adult body weight (Warham 1977; Lack 1968) the regression in Fig. 4 is plotted for 67 correlates representing 50 species of the order Procellarii- formes. The regression for egg mass, W = 0.7 13 fjw0.-3s , f = 0.98, X S E E = 1.17. For non- passerine birds (n = 557) a similar regression was reported (Rahn et al. 1985) where W = 0.399 BW'.''', f = 0.89, and XSEE = 1.48. 57 80 70 to 50 LO . -. .. . 30 - m.. 20 . . ,, . , -. . ~ 15 - - I 1 1 I I l l I I I I i l l l I 1 1 1 I 1 I1,x) . ' to too rn EGG WEIGHT ( g ) '9 Fig. 3. Regression of incubation time upon egg mass. Upper shaded region for 38 species of the order Procellarii- formes. Lower regression for 475 non-passerine birds. For details see text. (From Grant et al. 1982a). frocellariforrnes 'OoO 7 100 1 2. / t . , &Y ,," Yo' ' 0 735 W 10713 BW n 67 (50 species) r = ,992 SEE= ,067.X I 17 1 'I'o too 1000 10, Body Weight, g 30 Fig. 4. Regression ofegg mass upon adult body mass for 50 species (n = 67) of the order Procellariiformes. 58 While t h e e x p o n e n t s of t h e s e t w o regressions are n o t significantly different, t h e intercept f o r t h e Procellariiformes is (0.7 13/0.399), or I .8 times larger t h a n i n n o n - p a s s e r i n e birds. The relatively larger eggs o f this order h a v e b e e n described previously ( R a h n e t al. 1975; Rahn e t al. 1984b). We h a v e a l s o regressed a d u l t body mass against egg mass for t h e Procellariiformes using the same correlates as a b o v e . This regression is a d u l t body mass, BW = 1.73 W1.’4, 9 = 0.98, and XSEE = 1.23. Solving this e q u a t i o n for a n 83 g egg mass o f t h e Antarctic Petrel predicts an a d u l t body weight of 645 g, w h i c h can be compa- red with 641 g ( n = 34) given by Bierman & Voous (l950), cited by W a r h a r n ( l 9 7 7 ) , and 609 g ( n = 20) obtained in t h i s s t u d y . References Ar, A. & Rahn, H . 1980: Water in the avian egg: overall budget of incubation. Amer. Zool. 20,373 -384. Ar, A. & Rahn, H . 1985: Pores in avian eggshells: gas conductance, gas exchange and embryonic growth rate. Respir. Physiol. 61, I - 20. Bierman. W . H . & Voous, K . H . 1950: Birds observed and collected during the whaling expeditions of the Willem Barends in the Antarctic. 196-1947 and I947 - 1948. Ardea 37.1 - 123. Brown, D.A. 1966: Breeding biology of the Snow Petrel Pagodroma nivea (Forster). ANARE Scient$c Re- ports Ser. B ( 1 ) 89. Grant, G.S., Pettit, T.N., Rahn, H., Wittow, G.C. & Paganelli. C.V. I982a: Water loss from Laysan and Black-footed Albatross eggs. Physiol. Zool. -55, 405 - 4 I 4. Grant. G.S.. Pettit, T.N., Rahn, H., Whittow, G.C. & Paganelli. C.V. 1982b: Regulation of water loss from Bonin Petrel Plerodroma hypoleuca) eggs. Auk 99, 236-242 Lack, D. 1968: Ecological adaptations .for breeding in birds. Methuen, London. Mehlum, F., Bech, C. & Haftorn, S. 1985: Omithologi- cal investigations in Miihlig- Hofmannfjella, Dron- ning Maud Land. Pp. 27-34 in Orheim, 0. (ed.): Report of the Norwegian Antarctic Research Ex- pedition ( N A R E 1984185). Nor. Polarinst. Rapport. 22. I38 pp. Orton, M.N. 1968: Notes on Antarctic Petrels 7halas- mica antarctica. Emu 67.225-229. Paganelli, V.C., Olszowka, A. & Ar, A. 1974: ‘ h e Avian egg: Surface area, volume, and density. Condor 76, 3 19-325. Rahn, H., Paganelli, C.V. & Ar, A. 1975: Relation of avian egg weight to body weight. Auk 92.750-765. Rahn, H., Greene, D.G.,Taien, O., b o g , J . & Mehlum, F. 1984a: Estimated laying dates and eggshells con- ductance of the Fulmar and Brunnich’s Murre in Spitsbergen. OrnisScand. 15.110- 114. Rahn, H., Ackerman, R.A. & Paganelli, C.V. 1984b: Eggs, yolk and embryonic growth rates. Pp. 89- 112 in Whittow, G.C. & Rahn, H. (eds.): Seabird Energe- rics. Plenum Press, New York. Rahn, H., Sotherland, P.R. & Paganelli, C.V. 1985: Interrelationships between egg mass and adult body mass and metabolism among passerine birds. J . Om. 126.263- 271. Schonwetter, M. 1960: Handbuch der Oologie. Meise, W. (ed.). Akademie Verlag, Berlin. Tullett, S.G. & Board, R.G. 1977: Determinants of avi- an eggshell porosity. J . Zool. London 183.203-21 1. Warham, J . 1977: Wing loadings, wing shapes and flight capabilities of Procellariiformes. New Zealund J . Zool. 4.73-83. Watson, G.E. 1975: Birds of the Antarctic and Sub- Anrarcric. Amer. Geophysic. Union, Washington, D.C. Whittow, G.C. 1984: Physiological ecology of incubati- on in tropical seabirds. Pp. 47-72 i n Schreiber, R. (ed.): Tropical Seabird Biology. Studies in Avian Bio- log^^. No. 8. Cooper Ornith. SOC. Whittow,G.C.,Simons,T.R. & Pettit,T.N. 1984: Water loss from the eggs of a tropical seabird (Plerodroma phaeopygialat high altitude. Comp. Biochem. Physiol. 78A.537-540.