Egg cluster size variation in relation to the larval food

Document Sample
Egg cluster size variation in relation to the larval food Powered By Docstoc
					Res. Popul. Ecol. (1993) 35, 325--333.
~) by the Society of Population Ecology

                 F O O D A B U N D A N C E IN LUEHDORFIA PUZILOI
                          (LEPIDOPTERA: PAPILIONIDAE)

          Kazuma MATSUMOTO 1), F u m i n o r i ITO .2) and Yoshitaka TSUBAKI**

          Tama Forest Science Garden, Forestry and Forest Products Research Institute,
                              Todori-cho, Hachioji, Tokyo 193, Japan
       *Graduate School of Environmental Science, Hokkaido University, Sapporo 060, Japan
        **Laboratory of Wildlife Conservation, National Institute for Environmental Studies,
                                  Onogawa, Tsukuba 305, Japan


     Mean egg cluster size of Luehdorfia puziloiyessoensis varied among habitats. The
mean egg cluster size tended to be large when abundance of the larval food leaves
expressed as the fresh weight of leaves per unit area at a given habitat was high. Since
this variation was observed among closely located study plots (butterflies can easily
move between study plots), the egg cluster size variation among habitats is likely to be
a result of flexible response by females to varying food abundance for larvae.

    KEYWORDS: Luehdo~fiapuziloi, Papilionidae, egg cluster size, food abundance, oviposition.


     Luehdorfia butterflies deposit several eggs in clusters on leaves of their larval food
plants, wild gingers (Aristolochiaceae). The larvae in an aggregation consume all the
leaves of the original plant on which the eggs were laid, and then disperse in search for
new plants. The beneficial effects of aggregation in young larval stages (mortality rate
reduces probably due to feeding facilitation among larvae) may favor larger egg
clusters (Osada, 1983; Matsumoto, 1989, 1990), whereas clustering of too many eggs
results in earlier consumption of the original plant, causing high mortality during
the subsequent dispersal stage (Fuzisawa et al., 1964; Ikeda, 1976; Ito et al., 1982;
Matsumoto, 1990). In L. japonica, mean egg cluster size varies in different local
populations (Ishida, 1982) and it is correlated with the amount of available larval food
at each habitat, suggesting that female butterflies control egg cluster size to maximize
the survival of their progeny (Tsubaki, in press).

  1) Present Address: Tropical Agriculture Research Center, Ohwashi 1-2, Tsukuba 305, Japan
  2) Present Address: Faculty of Education, Kagawa University, Takamatsu 760, Japan

     The above observations suggest that egg cluster size may partly be a result of the
trade-off between cost and benefit of egg clustering. However, because L. japonica
depends on the locally diverse wild gingers, Heterotropa spp., the egg cluster size may be
different in different populations because each population utilizes different host plant
species. Therefore, there are at least three possibilities how egg cluster size is
determined. First, local variation in egg cluster size may be genetically controlled, or
second, it could be due to facultative decision making by individual females which may
be possible after gathering information of host plant availability. Finally, differences
in nutrient composition among host plant species may be responsible for the egg
cluster size variation: eg., butterflies ate more nutrient-rich host plants during their
larval stage may lay eggs in larger clusters than the others. Thence, we carried out a
field study on a population (composed of several sub-populations among which adult
butterflies can easily migrate) of a sibling species, L. puziloiyessoensis that depends on a
single host, Asiasarum heterotropoides. In this population, we assumed to be able to
eliminate both of the effects of host plant species and genetic differences among sub-
populations on the egg cluster size, allowing us to test whether females modify their
egg cluster size in response to the abundance of larval food resources.

                               MATERIALS AND M E T H O D S

      Egg cluster size variation and food leaf abundance were measured for the Hok-
kaido subspecies, L. puziloiyessoensis,and itshost, Asiasarum heterotropoides,
at three study plots (Plots i, 2 and 3) in a western suburb of Asahikawa City in early
June, 1990, and at a plot in the Tokyo University Forest in Furano City, Hokkaido
(Higashiyama plot), in late M a y 1988-89. The Asahikawa study area was a decidu-
ous forest and the food plants were commonly seen on the ground. Higashiyama Plot
is a mosaic made up of bushes, deciduous trees, and a plantation of larch, and the food
plant at this plot occurs as a small isolated patch. In general, the butterfly and the
food plant were rare in the University Forest, and the Higashiyama Plot is the only
area where we found a sufficientnumber of egg clusters for statisticalanalysis.
     At the above four study plots, the sizes of egg clusters were recorded. The food
plant distribution was not uniform and the eggs tended to occur where the plant was
abundant. The representative density of food plant leaves at each plot in Asahikawa
was measured by setting five 1 • 1 m quadrates along five 5 m transects (25 quadrates
in total). As we avoided areas with few host plants in this sampling procedure,
the density of host plants should be referred as the "density within a patch". At
Higashiyama plot, we set a 22 • 28 m rectangle study site (616 quadrates of I m 2)
including most of host plants occurring in the area (in order to study the survival of
larvae; the results will be reported elsewhere) in 1989. W e chose five 5 m transects
(consecutive 5 quadrates) (25 quadrates in total) for the estimation of leaf density avoid-
ing areas with few host plants. This procedure may, more or less, standardize the

comparison of host plant density between the Asahikawa and Higashiyama plots.
Approximately 100 leaves of the food plant in and/or near the quadrates were col-
lected. They were brought to the laboratory in a plastic bag to avoid desiccation, and
their fresh weight was measured by a semi-micro balance within a day of collection.


Variation in Egg Cluster Size among Populations
      Figure 1 shows frequency distributions of the egg cluster size of L. puziloi
yessoensis. The average number of eggs varied between 8.3 and 12.0. There was no

                                   ~,sahikawa Plot 1
         3O                                    n=21


          4O                                Plot 2                       50              Higashiyama 1988
                                   -----] n=25                           40                                 n =44

                                                                                                      I     mean=lO.3_+4.3

 t-       20
          10                                                             10
           0'      I                                                      0

          70                                                             50i
                                                                                         Higashiyama 1989
          60                                          Plot 3             40
          5O                                                                                              n=28
                                                                         30                                 ean=9.7+~.7
          40                                     n=8
          30                                     mean=l 2.0!-_2.6
                                                       I   I     I                                                    !         !   !
           0 ~         co                                                 0    co co o~
                            r...                                               ,--
                                                                               9     ~   r'..
                                    o    03      co
                                    .r   T-.                                                    T--       T--   ~--       T--

                                                               Egg cluster size
        Fig. 1. Frequency distributions in the egg cluster size of L. puziloi at the three plots in Asahi-
             kawa and one plot in Higashiyama.

               14                               m
       t./)    1G
       co)       8          000                                     O0                          9
                                         I          I           I        I        I         I           I         I
                  0            5        10 15 20 25 30 35 4O                                                      45
                                         Number of host plant leaves / plant

       N 10              D 9                            9
                      ,9               9149             9
                      l l                               9
      o          6
                 2             9
                                   I           I            I                I         I            I                 I

                     0             20         40            60               80       100       120               140
                                             Number of host plant leaves / m2
      Fig. 2. Relationship between host plant size (number of leaves) and egg cluster size laid on it
           (a), and relationship between host leaf density (number of leaves/m 2) and egg cluster size

significant difference in egg cluster size between the two years at H i g a s h i y a m a Plot
( M a n n - W h i t n e y U test, z = 0.47, ns). Significant differences were detected a m o n g the
four plots (Kruskal-Wallis test, H = 1 1 . 8 1 ,          P < 0 . 0 1 ) and a m o n g the three plots in
Asahikawa only (Kruskal-Wallis test, H = 11.12, P < 0 . 0 1 ) .

Egg Cluster Size and Food L e a f A b u n d a n c e
    Figure 2a shows the relationship between egg cluster size and host plant size
( n u m b e r of leaves) at H i g a s h i y a m a plot in 1989.    W e could not find any relationship
between t h e m (r 2 = 0.058, P > 0.2).        T h e n we tested whether egg cluster size is associ-
ated with food leaf density ( n u m b e r of leaves per 1 m 2) at H i g a s h i y a m a plot (Fig. 2b).
Again, we could not find any relationships (r2=0.012, P > 0 . 5 ) .

Local Variation in Food L e a f A b u n d a n c e
    T a b l e 1 shows flesh weight and density of the leaf of A. heterotropoides in the four
plots in Asahikawa and H i g a s h i y a m a .     W e see that both of leaf density and leaf weight
differ from habitat to habitat.         T h e difference in the leaf weight is due to the difference
in leaf area, but the reason of this local difference is not clear.              Soil type, soil moisture
content, nutrient conditions, s u r r o u n d i n g vegetation etc. m a y be responsible for the
     Figure 3 shows the relationship between m e a n food leaf density and m e a n egg
cluster size at each plot. A Value from H i g a s h i y a m a 1988 is shown by an open circle
to distinguish from others, as we have no host plant data this year. This point was
plotted assuming that there is no difference in the host density between 1988 and
1989.      It is known to be little annual change in host plant density in several species
of ginger Heterotropa spp. ( M a t s u m o t o 1990, T s u b a k i unpublished).           T h e m e a n egg
cluster size was positively correlated with the m e a n food leaf density when all of study
plots are included (P = 0.85, P < 0.05, n = 5), and just not significant when data from
H i g a s h i y a m a in 1988 is excluded from the analysis (r2=0.85, P = 0 . 0 7 , n = 4 ) .
      W h e n we regressed the m e a n egg cluster size to the average a m o u n t of leaf in
weight per square m e t e r (average leaf density multiplied by m e a n leaf weight) at each

  Table 1. Fresh leaf weight and leaf density of Asiasarum heterotropoides at four study plots in
                                        Leaf weight (mg)                        Leaf density (per m 2)
                                   n          Mean          SD             n*          Mean          SD
  Asahikawa (1990)
    Plot 1                         93         707.7        303.1           25            6.9          7.0
    Plot 2                        107         817.9        366.4           25           39.3         24.0
    Plot 3                        103        1036.8        369.0           25           34.9         29.2
  Furano (1989)
    Higashiyama                   113         545.5        243.2           25           28.2         33.3
  * Number of quadrates (1 x     1 m)


  N                                                                               9     I
  r                                                                               '     .I.
                                                          1988                          T PIot 2
  "-'t      10                                                   A
  ~O                                                         T
                               Pot 1                        Higashiyama

                         I        I        I       I        I        I        I             I      I         |

                0               10                20                 30                 40                  50
                                Host leaf density (number/m2)
         Fig. 3. Relationship between the mean egg cluster size and mean host leaf density at the three
              plots in Asahikawa and at Higashiyarna Plot. Vertical and horizontal bars indicate 95%
              confidence limit intervals of the mean.

locality, we got a higher correlation coefficient (rZ=0.953, P < 0 . 0 1 , n = 5 ) .                   This m a y
suggest that females are assessing the a m o u n t of food leaves (area or volume) rather
than leaf density ( n u m b e r ) .


         T h e present study showed that the m e a n egg cluster size of L. puziloi varied from
 place to place and that the m e a n egg cluster size was positively correlated with food
leaf a b u n d a n c e at a given habitat. It is especially notable that these p h e n o m e n a were
found a m o n g the three very closely located plots in Asahikawa.                      Although the adult
dispersal range of L. puziloi has not been studied so far, a study on L. japonica
( M a t s u m o t o , 1984) showed that the adults of both sexes will disperse over a kilome-
ter. T h e distance between Plots 1 and 2 was only a b o u t 100 m , a n d those between
Plots 1 and 3 and between Plots 2 and 3 were both about 300 m.                        As the flight ability of
L. puziloi does 'not seem to be largely different from that of L. japonica, these plots were
p r e s u m a b l y n e a r enough to one another to allow frequent gene exchange b y the resi-
dent butterflies.       Accordingly, all eggs found in Asahikawa plots p r o b a b l y belong to a
single deme. This suggests that females modify their egg cluster size in response to the
a b u n d a n c e of larval food plants.

       In several butterfly species, females are known to be able to detect conspecific eggs
 and make a choice to oviposit or not on a given plant to avoid competition among
 larvae (Rausher, 1979; Shapiro, 1981). Females m a y respond to egg-load on host
 plants and adjust egg cluster size. However, we feel it unlikely in Luehdo~fia butterflies
 with several reasons mainly based on observations on L. japonica. First, both of L.
japonica and L. puziloi females appear to have no particular behaviour in search for
 conspecific eggs which is reported in Buttus philenor females (Rausher, 1979). Second,
 we often found several clutches together (up to seven in L. puzilo 0 on a host leaf and
 spatial distribution of clutches is highly clumped (on L. japonica by Matsumoto, 1990,
 and L. puziloi by Matsumoto, unpublished). Finally, a comparison of clutch size
 showed no difference between the first and the second clutches laid on a same host leaf
 (on L. japonica Tsubaki, unpublished). Therefore, we presume that the effect of egg-
 load assessment on the clutch size is negligible in Luehdorfia butterflies, if any.
       Then, how female butterflies can assess the abundance of food plants? When L.
 puziloi females are trying to oviposit, they fly slowly near the ground, and frequently
 alight on green objects including host and non-host plants. Like other Papilionid
 butterflies, L. puziloi females are probably using chemical sensory organs distributed
 on the tarsi of forelegs to identify correct food plants. In L. japonica, Tsubaki (in press)
 found that females scarcely lay eggs immediately after alighting on a correct host plant;
 instead, they fly up again to search for another one (usually within l m of the first host
 plant). Females repeat this behavior for m a n y times (usually 10-50 times), and then
 they finally start to lay eggs on the last host plants on which they alighted, or fly away
 without oviposition. L. japonica females tended to lay eggs only when they could alight
 on host plants frequently, where host plants were densely occurred. We do not
 have yet any precise record of host seeking behaviour of L. puziloi, but it appears to be
 similar to that of L. japonica (Kurakane, 1983; Hara, 1984, and personal observation
 by Matsumoto). Therefore, L. puziloi females probably be able to estimate host plant
 density, and then adjust their egg cluster size accordingly.
       The above observations, however, do not exclude a possibility that female
 butterflies are able to assess the quality of host leaves. We showed that the mean
 clutch size was small where host leaf density was low in Fig. 3. If the nutrient condi-
 tions or water content of soil, surrounding vegetation and other environmental factors
 at each site affect both the density and quality of host leaves, the positive relationship
 m a y also be found between clutch size and quality of leaves. Chemical analyses of
 host leaves are needed for further discussion about the host plant assessment by
 Leuhdorfia butterflies.
       Egg cluster size adjustment by swallowtail butterflies in response to the quality of
 individual host plant has been reported by Pilson and Rausher (1988) in Buttus
 philenor. In the case of L. puziloi, however, we could not find any relationship be-
 tween the egg cluster size and individual host size or food abundance in a square
 meter. Instead, we found a positive correlation between egg cluster size and host

d e n s i t y i n a local h a b i t a t .   Size of a r e a i n w h i c h female butterflies s h o u l d e s t i m a t e the
food a b u n d a n c e for l a r v a e m a y be related to the ability of l a r v a e to m i g r a t e f r o m o n e
host p l a n t to a n o t h e r .     I k e d a (1976) r e p o r t e d t h a t L.   puziloi l a r v a e m i g r a t e d u p to t e n
m e t e r s after e x a u s t i o n of the o r i g i n a l host.         T h e r e f o r e , L. puziloi females m a y be
a d j u s t i n g t h e i r egg cluster size i n r e s p o n s e to the food a b u n d a n c e i n a n area s o m e w h a t
l a r g e r t h a n a s q u a r e m e t e r b u t s m a l l e r t h a n the whole area of a h a b i t a t .

      ACKNOWLEDGMENTS: We are thankful to members of the Tokyo University Forest in Hokkaido for
their assistance, especially to Professor S. Watanabe, the chief, for allowing the field study at Higashiyama
with kind consideration and to Mr. H. Arisawa for valuable support in many aspects of this study. We
thank Dr. Perri Eason for her comments on the manuscript and suggestions on English expressions. This
study was supported in part by the Grant-in-Aid for Scientifc Research from the Ministry of Education,
Science and Culture of Japan (No. 63790356, and 05454011) awarded to KM and YT, respectively.


Fuzisawa, S., M. Kinoshita, B. Kitamura, M. Kurata and H. Mizukami (1964) Luehdorfiajaponica and
      L. puziloi inexpecta.          Shinano-Kyoiku-Kai, Nagano. (in Japanese)
Hara, S. (1984) Luehdorfia japonica in the Himekawa valley. Tsukiji-shokan, Tokyo. (in Japanese)
Ikeda, K. (1976) Bioeconomic studies on a population of Luehdorfia puziloi inexpevta Sheljuzhko
     (Lepidoptera: Papilionidae). Jap. J. Ecol. 26: 199-208.
Ishida, T. (1982) On the local variation of egg cluster size in Luehdoo~ajaponica. Iratsume, 6: 10-18. (in
Ito, Y., Y. Tsubaki and M. Osada (1982) Why do Luehdorfia butterflies lay eggs in clusters? Res. Popul.
     Ecol. 24: 375-387.
Kurakane, O. (1983) Luehdorfia japonica and L. puziloi. Bunseid0-Shinkousha, Tokyo. (in Japanese)
Matsumoto, K. (1984) Population dynamics of Luehdorfiajaponica Leech (Lepidoptera: Papilionidae). I. A
    preliminary study on the adult population. Res. Popul. Ecol. 26: 1-12.
Matsumoto, K. (1989) Effects of aggregation on the survival and development on different host plants in a
      Papilionid butterfly, Luehdorfiajaponica Leech. Jap. J. Ent. 57: 853-860.
Matsumoto, K. (1990) Population dynamics of Luehdo~ajaponica Leech (Lepidoptera: Papilionidae). II.
    Patterns of mortality in immatures in relation to egg cluster size. Res. Popul. Ecol. 32: 173-188.
Osada, M. (1983) Meaning of egg-mass and larval aggregation in Luehdoo~ajaponica Leech. Konchu to
    Shizen (Nature and Insects) 18(5): 2-5. (in Japanese)
Pilson, D. and M. D. Rausher (1988) Clutch Size adjustment by a swallowtail butterfly.                               Nature 333:
Rausher, M. D. (1979) Egg recognition: its advantage to a butterfly. Anim. Behav. 27: 1034-1040.
Shapiro, A. M. (1981) The pierid red-egg syndrome. Am. Nat. 117: 276-294.
Tsubaki, Y. (1994) Clutch size adjustment by Luehdo~fiajaponica. In M.J. Scriber, Y. Tsubaki and R.
    Lederhouse (ed.) Ecology and evolutionary biology of the Papilionidae. Scientific Publishes Inc. and John
    Heppner. (in press)

Shared By: