estacionalidad de insectos.pdf

8/16/2019 estacionalidad de insectos.pdf

1/15

Seasonality of Tropical Insects

Author(s): Henk Wolda

Source: Journal of Animal Ecology , Vol. 49, No. 1 (Feb., 1980), pp. 277-290

Published by: British Ecological Society

Stable URL: http://www.jstor.org/stable/4289

Accessed: 10-05-2016 23:29 UTC

R F R N S

Linked references are available on JSTOR for this article:

http://www.jstor.org/stable/4289?seq=1&cid=pdf-reference#references_tab_contents

You may need to log in to JSTOR to access the linked references.

Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at

http://about.jstor.org/terms

JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of content in a trusted

digital archive. We use information technology and tools to increase productivity and facilitate new forms of scholarship. For more information about

JSTOR, please contact [email protected].

British Ecological Society, Wiley are collaborating with JSTOR to digitize, preserve and extend access toJournal of Animal Ecology

This content downloaded from 170.210.197.63 on Tue, 10 May 2016 23:29:20 UTCAll use subject to http://about.jstor.org/terms


2/15

Journal of Animal Ecology (1980) 49, 277-290

SEASONALITY OF TROPICAL INSECTS

I. LEAFHOPPERS HOMOPTERA) IN LAS CUMBRES, PANAMA

BY HENK WOLDA

Smithsonian Tropical Research Institute, P.O. Box 2072, Balboa, Panama

SUMMARY

(1) The seasonal distribution of tropical insects was studied using a light trap in Las

Cumbres, Panama, over a period of 3- years.

(2) The phenology-measures Seasonal Range (SR) and Seasonal Maximum (SM) are

used here for the first time.

(3) Some 370 % of the species occur throughout the year. But almost all species, with

few exceptions, do have clearcut seasonal peaks in abundance.

(4) The results are compared with data on Homoptera from Finland, England and the

USA. Seasons as short as those commonly found as far north as Northern Finland also

occur in the wet tropics of Las Cumbres, Panama.

(5) Most species have their seasonal peak in abundance some time during the rainy

season. The concentration of peaks in the early dry season (January-February) is caused

by increased activity of species leaving the grass that is drying out, in search of alternative

food.

(6) Some of the seasonal peaks occurring at the very beginning and at the end of the

rainy season refer to dispersing insects, including some species which do not normally live

in the vicinity of the trap.

INTRODUCTION

The occurrence of seasonal fluctuations in the number of individuals in tropical insect

species is a well established fact (for references see Wolda 1978a, 1979a). However, it is

far from clear how such fluctuations compare with those found in other parts of the

world with different climatic regimes. On the basis of climatic data one might expect

tropical species to have longer seasons than their counterparts at higher latitudes. Many

species may be expected to occur around the year, or at least throughout the favourable

season, which usually is the rainy season. One might expect many of the species that do

occur around the year to have little, if any, seasonal variation in abundance, as shown for

instance for some butterflies (Ehrlich & Gilbert 1973; Owen & Chanter 1972). However,

is this true? The published data on seasonality in the tropics are insufficient to allow for a

general test of these, or any other, hypotheses concerning the seasonality of tropical

insects.

For such a test, the seasonal distribution of a large number of tropical species would

have to be documented and these data would have to be analysed quantitatively so that

direct comparisons can be made between different areas in the world. The present paper

will present such an analysis for Homoptera from Las Cumbres, Panama.

0021-8790/80/0200-0277 02.00 (?1980 Blackwell Scientific Publications

277



3/15

278 Seasonality of Homoptera in Las Cumbres, Panama

The insects are collected with a light trap. I believe that for many species these data

reflect reasonably well the real fluctuations in abundance in nature, that is of the active

adults of those species. In some instances I do have general observations supporting this

contention, such as for cicadas, where monitoring the songs gives an impression of

seasonal presence which coincides nicely with the light trap data. For a further discussion

see Wolda 1978a, b.

Attempts have been made in the literature to capture some aspect of the seasonal

distribution of a species in a single parameter. Slobodchikoff & Parrott (1977) used a

descriptor called 'Seasonal Niche Breadth' (Levins 1968). Unfortunately, this measure is

useful only when the information on phenology for each species is based on a large

number of individuals (Wolda 1979b). For smaller sample sizes the measure is practically

useless. I used 'Season Length' (Wolda 1977), which is much better than the Seasonal

Niche Breadth, but still dependent on sample size (Wolda 1979b). I proposed a measure

called 'Seasonal Range' (SR), which is an estimate of the length of the season in which the

phase of the insect under consideration, in this case adults, are present in a given year,

corrected for sample size (Wolda 1979b). This measure SR will be used together with a

new one called 'Seasonal Maximum' (SM) to analyse the light trap data on Homoptera

from Las Cumbres, Panama and to compare them with data from Kansas, USA (Blocker,

Harvey & Launchbaugh 1972 and personal communication), from England (Waloff

1973) and from Northern Finland (Kontkanen 1950).

PROCEDURES

Locality and trapping

The data to be discussed were collected using a light trap of the Pennsylvania type,

modified for the tropics by my colleague Nicholas Smythe. In this trap the insects are

attracted by a Sylvania Fl 5T8 BL fluorescent light, fall through a funnel in a cloth bag

where they are killed by carbon tetrachloride. The trap was in operation all night, every

night, from October 1973 to March 1977. The insects were either sorted immediately or

stored in a freezer for later sorting. Several groups of insects have been sorted out and

many of these have been analysed by specialists to the species level. The present paper only

deals with the Homoptera, sorted into species by my assistant Mr Miguel Estribi and

myself. Other groups will be dealt with elsewhere.

For the vast majority of the homopteran species the classification should be correct.

We received considerable help from various specialists when taxonomic problems needed

to be resolved. It is possible that in some cases we may have to split, or lump, species as

the taxonomy of the groups is re-examined. For instance, the small green Empoasca

(Typhlocybinae) are all counted as just one taxon although there probably are several

species represented.

The trap was located in Las Cumbres, a town some 15 km North of Panama City,

Republic of Panama, at 9?5'36 N and 79?3 '54 W. The trap was on top of a slope. South

and west of the trap there are some bungalows, surrounded by trees, shrubs and lawns.

To the east and north the trap overlooks a valley with many wild, ornamental and fruit

trees. Fifty metres in that direction is the edge of a young second growth forest. The

elevation is 150 m above sea level.

Mean annual rainfall is 1500 mm, 84% of which falls from May to November and 900%

from April to November. The dry season usually starts some time in December and lasts



4/15

H WLDA279

30- a)

20

I0

b)

400

300

200

100

-112 I 1 l l llL1 II IF t I,

1974 1975 1976

FIG. 1. (a) The number of rainy days per month in Las Cumbres, Panama, and (b) rainfall

in mm during the study period.

until April or May. Towards the second half of January the effect of the dry season begins

to show in that the lawns and roadside grasses begin to dry and turn brown. The rainfall

per month for the study period, together with the number of days with rain per month, is

plotted in Fig. 1 (data from Caballero 1974, 1975, 1976, 1978). No data are available yet

from that source for 1977, but only little rain fell in January 1977 and February and

March of that year were completely dry.

The mean temperature per month is 27 ?C, with no apparent seasonal fluctuation. The

mean maximum temperature is about 31 ?C with the maxima in March and April, the

last months of the dry season, generally somewhat higher than during the rest of the year.

The mean minimum temperature is 230 C and tends to be somewhat lower in the dry

season than during the rainy season.

Measures of seasonality

Rather than trying to capture many aspects of the seasonal distribution in one single

parameter, as the Seasonal Niche Breadth pretends to do, I selected two aspects of the

seasonal distribution and tried to represent each by a parameter. This procedure seems

less confusing. The aspects selected are the length of the time, per year, that a species is

present and the relative height of the seasonal peak. These two aspects are not completely

independent, but sufficiently so to warrant two parameters instead of just one. For an

area where all species have short seasons, one parameter would be sufficient, but for

species which occur around the year it can be very important to be able to distinguish

between species which have a very sharp seasonal peak and those which show no seasonal

variation in abundance at all.

The Seasonal Range (SR) is determined by measuring the length of the season (SL)

in weeks. With the help of a simulation model based on a seasonally normal distribution,

the SL is transformed to SR, thus eliminating, at least to some extent, the effect of sample

size. For details on how the transformation from SL to SR is carried out, see Wolda

1979b).

The Seasonal Maximum (SM) is determined as follows. In a year there are 13 periods of

4 weeks. The mean number per 4 weeks is determined simply by dividing the annual



5/15


total by 13. The running 4-week sum of the data is taken and the 4-week period with the

highest number of individuals is called the maximum. Seasonal Maximum is this maximum

number divided by the mean. Seasonal Maximum varies from 13, when all individuals

are found within one period of 4 weeks, to one, when the number of individuals is abso-

lutely constant throughout the year. For actual data, however, because of statistical

variations both in the actual populations and in the sampling process, even in species

which do not have any biologically significant seasonal variation in abundance, SM will

always be larger than unity.

As one might expect, SM is also dependent on sample size. This can be demonstrated

by taking computer-drawn samples of various sizes from seasonally normal distributions,

with overlapping tails, with various seasonal standard deviations (SSD). An SSD of 1

means a very sharp seasonal peak, an SSD of 20 describes a distribution which is effectively

rectangular, i.e. non-seasonal. For further details of these distributions see Wolda 1979b.

For each SSD and each sample size ten replicate samples were obtained, which gives a

mean and a standard deviation for SM. For a number of SSD values the mean value of

SM is plotted against sample size in Fig. 2. The lines given are actually lines drawn by

eye through the calculated points. The peculiar bump in the line for SSD = 2 could be a

sampling error. For anyone who wishes to use these same lines the data describing them

are available on request. For smaller samples the effect of sample size is considerable.

Actual data would be represented by points scattered over the area of Fig. 2 and the lines

classify the points in groups, such as the ones above the line for SSD = 1, the ones

between the lines for SSD = 1 and SSD = 2, etc. The distribution of the points over

these groups describes the data set in terms of the sharpness of the seasonal peak,

eliminating, at least to a certain extent, the effect of sample size. The groups associated

with the lines for low values of SSD indicate sharp seasonal peaks, the ones with high

values for SSD refer to less pronounced seasonal variation.

Admittedly, the lines in Fig. 2, which are used for all data sets, are based upon a

seasonally normal distribution, while an actual distribution may deviate from normality,

sometimes even considerably so. However, I doubt that such deviations from normality

affect the shapes and position of the lines in Fig. 2 in any serious way.

r w _ _ I Z 155~~~~~~~5D

1

E_

.E_

I 0

02

C :

0

5 - 4

U)

7

20

10100100010000

Sample size

FIG. 2. The relations between the Seasonal Maximum (SM) and sample size in computer-

drawn samples from seasonally normal distributions with standard deviations (SSD)

varying from 1 to 20.



6/15

H WLDA281

RESULTS

Species richness

A total of 86 039 individual Homoptera were caught, representing about 540 species.

The distribution of the Homoptera over the various families is given in Table 1, with the

Cicadellidae given per subfamily. Some 750 of the individuals and about 5000 of the

species are Cicadellidae, as seems to be the case in more open environments in Panama.

In forests Fulgoroidea are usually more abundant. About 2000 of the species can be

classified as common, i.e. they were caught with a hundred individuals or more, while

about the same percentage (107 species) is represented by one single individual only. This

shows that after 31 years of intensive trapping many species present have not been

caught at all. The number of 540 species represents only a minimum figure for the total

number of species within reach of the trap.

TABLE 1. Homoptera collected with a light trap in Las Cumbres, Panama, over

a period of three and a half years

Common

species

Individuals Species N 2 100

caddae 1834 105

Mmbracidae 231 30

Crcopdae 83 11

sylidae 9315

Ccadellinae 19118 35 10

galliinae 6066 104

Xestocephalinae 9595 11 6

Hcaline 11

rvannae 4 2

Ledrne 2 1

Deltocephalinae 8834 70 19

Typhlocybnae 13375 89 17

docernae 1168132

yponnae 4903335

Nocoelidiinae 66 5

oeidinae 134

hlidae 45982110

Cxde 221832 7

Aanalonidae 18 2

hlixidae 11011

ugordae 602

Nogodindae 21 1

nnardae 15 3

Fatde 3294 23 4

erbdae 335056 5

Delphacidae 6649 47 14

Tropduchdae 6 2

Dictyopharidae 314 10 1

otal 86039540110

Seasonal maximum and seasonal range

Some examples of seasonal distributions are given in Fig. 3. These range from a very

short season in Amblyscarta resolubris (Fowler) to a virtually constant distribution in

Polana scinna DeLong and Freytag. All these examples have their maximum, if there is one,

in the wet season. In Fig. 4 some aberrant species are given. Pacarina signifera Walker, a

common cicada, is typically a dry season species. Muirolonia metallica Fowler has a

bimodal distribution while Pintalia sp. 62B has three peaks per year.



7/15


10_ a)

O 0S

25 b)

o7 ~~~~~~

0 LLA

100 _ (d)

100 e)

1974 1975 1976 77

FIG. 3. The seasonal distribution, in numbers per 4 weeks, for some species of Homoptera

in Las Cumbres, Panama. (a) Amblyscarta resolubris (Fowler) (Cicadellinae). (b) Myconus

uniformis Metcalf) Achilidae). c) Muirilixius banksi Metcalf Achilixiidae). d)

Xestocephalus tessellatus Van Duzee Xestocephalinae). e) Polana scina DeLong and

Freytag (Gyponinae).

For each year, every species which is represented by at least six individuals is used in the

analysis. There are 3 complete years, so that each species may contribute up to three data

points. The 246 species used give a total of 637 data points.

The values for SR are plotted, per group, in Fig. 5. A total of 235 points, 36.9%, are

52, which means that the species occurred the year around. Many species have shorter

seasons, or even very short seasons, much shorter than one might expect from the

climate. It is just possible that in some cases this is because the species does not live near

the trap and is only caught on its annual dispersal flight. I cannot rule out this possibility.

Most known dispersers that do not live near the trap are caught twice per year, making

the season rather long instead of short, once around May and once towards the end of

the year. In many cases, however, the shortness of the season is undoubtedly real.

25

t 25 - (b)

0r-

w50-(c) A

974 1975 1976 77

FIG. 4. The seasonal distribution, in numbers per 4 weeks, of some species of Homoptera.

(a) Pacarina signifera (Walker) (Cicadidae). (b) Muirolonia metallica Fowler (Cixiidae).

(c) Pintalia sp. '62B' (Cixiidae).



8/15

H WLDA283

10 (a)

N=17

n 59

10 -(b)

_N134

I I I rr1 .- 30 8

10 ( c) N= 3 (unshaded)

_ N =2 shaded)

F r N- 188 36.7

20 - d)

F N= 69 ilrl flYEfltfKFl.,.r

N= 637

0~~~~~~~~~~~~~~~5

10 (e)2023

01020o/5

Seasonal Range

FIG. 5. Distribution of the values for the Seasonal Range (SR) in some groups within the

Homoptera. a) Cicadidae, b) Membracidae, c) Cercopidae: unshaded) Psyllidae:

(shaded), (d) Cicadellidae, (e) Fulgoroidea-Delph.; (f) Delphacidae, (g) Homoptera total.

The frequency of species occurring throughout the year is given in numbers and percent-

ages rather than in histograms. Data from Las Cumbres, Panama.

The 637 values for SM are plotted against sample size in Fig. 6. The size of each point

is directly proportional to the number of data points it represents. The lines are explained

above. The distribution of these data points over the spaces between these lines is given,

in percentages, in Table 2. There is a wide range of values, ranging from very sharp peaks

to no peaks at all, as illustrated in Fig. 3. Most cases are seasonal distributions which

resemble normal distributions with a seasonal standard deviation between 4 and 11. There

are a few species which do not show any seasonal variation in abundance, but those are

in the minority, even among the species that are present as active adults around the year.

Several species have well-defined sharp seasonal peaks. Among these there are some

where this peak is an increase in activity, rather than a maximum in abundance. This is

the case for some species which live in the lawns and start dispersing in search of suitable

places to live in the early dry season. Good examples of this are Plesiommata mo/lice/la

(Fowl.) (Cicadellinae) and A gal/ia modesta Osb. and B. These, and some other similar

species make up most of the points in the upper right hand corner of Fig. 6. In many

cases, however, the sharp peaks undoubtedly represent such sharp peaks in abundance

in the real world.

There tend to be differences between families in the average seasonal distribution of

their species (Fig. 5; Table 2). Cicadas have a rather short season, which is confirmed by



9/15


_- - I I 155D

10 a

E

E

En

E* . a ... :- .-.. * 7

5=p4

20

C

10100100010000

Sample size

FIG. 6. Plot of values for Seasonal Maximum (SM) v. sample size for Homoptera from

Las Cumbres, Panama. The size of each plot is proportional to the number of data points

it represents. The simulation lines of Fig. 2 are also included.

TABLE 2. Seasonal Maximum (= sharpness of the seasonal peak in abundance)

for leafhoppers (Homoptera) from Las Cumbres, Panama as compared with

Homoptera from Kansas, England and Finland. The distribution of the species

over the various categories is given in percentages. The categories range from

a very sharp seasonal peak on the left to the virtual non-existence of any

obvious peak on the right. The number of data on which the percentages are

based are given (N). For further details see text

SSD-classes of SM< 1 1-2 2-4 4-7 7-11 11-20 > 20 N

Las Cumbres, Panama

caddae 5-9176 47235 5917

Mmbracidae - 77 308 231 15 4 23-1 13

Crcopde 66733 33

Pyde - 100-

Ccadellidae 0-6 3-8 116 36-2 296 15-4 2 9 345

Fulgoroidea-Delphac. 1 6 12.2 46-3 30 8 64 2-7 188

Delphacidae 43 319 42 11-6 5-8 4-3 69

Total Las Cumbres 0.5 4.1 14-7 39.1 27 11-3 3-3 637

Kanas, USA3-8 25 40-4 23-1 5-8 19 - 52

ngand9-132743-6 1271-855

Fnand232 581163 23 43

data on their singing. The one species listed here as having an SR of 52 I tend to interpret

as an error in estimation. In 1975 the species Herrera marginella (Walker) was caught

with 6 individuals only over a period of 25 weeks. Using my standard procedure (Wolda

1979b) this gives an SR of 52. With samples this small some errors are bound to appear.

I think this is one instance in which the 25 is real, the 52 is not. The SR values are grouped

in classes 0-10, 10-20,. . ., 40-51.9, and 52 and differences between groups are tested by

x2. The same test is used to compare groups in the distribution of the SM values. The

Cercopidae and Psyllidae are left out of the comparisons as the number of data points is

too low. The differences between Membracidae. Cidadellidae and Fulgoroidea (Minus

the Delphacidae) are not significant, or only barely so at the 5%o level. The Delphacidae



10/15

H WLDA285

13 -

5

I0-

E

- 26

C

C~~~~55

0

127

189

235

l I I I

1020304050

Seasonal Range

FIG. 7. The relation between SR and SM for the Homoptera from Las Cumbres, Panama.

Horizontal lines indicate the range of SR values used for calculating a given point, and the

mean value of SM for those data. The vertical lines are at the mean value of SR for that

subset of data and indicate one standard deviation in each direction. The numbers indicate

the number of data points represented.

have significantly shorter seasons and significantly sharper seasonal peaks, on the

average, and the Cicadas again have significantly shorter seasons, even than the Del-

phacidae, and sharper seasonal peaks.

As mentioned before, SR and SM, although far from identical, are correlated measures

of seasonality. For the 637 Homoptera data points the relation between SR and SM is

shown in Fig. 7. The SR values are grouped as indicated by the horizontal lines. These

lines are drawn at the level of the mean value of SM for these data (SM here is not

corrected for sample size). At the mean value for SR within each group, a vertical line

indicates one standard deviation of SM in each direction. The numbers indicate how

many data points there are in each group. There is a clear relation between SR and SM,

but the fairly large standard deviations in SM indicate that the knowledge of only one of

the two parameters has only limited predictive value for the value of the other.

An important problem in an analysis such as this is presented by those species that are

multimodal in distribution. Some of the more common species show two or more well-

defined peaks in abundance per year (Fig. 4). In other cases there could be more than

one peak but the picture is not entirely clear. Some of the less common species seem to

have more than one peak, but the interpretation of such data is rather subjective. In

some cases the light trap data simply are not good enough to demonstrate the number of

peaks (generations) per year. A good example is Aeneolamia postica (Walker) (Cer-

copidae). It was caught in the present trap only occasionally, but at the lights of a house

nearby, on a wall facing a grassland where the species occurs in large numbers, the species

is very common. The distribution per day is given in Fig. 8, indicating some six or seven

peaks during the rainy season. Such a resolution can only be obtained for species that

are very common in the trap. The maxima in abundance shown in Fig. 8 are real and not

an artefact produced, for example, by phases of the moon.



11/15


300

200

I oo9

50

2 5

0- --7 II ~ ~ ~ J~~?iL l

vZ a I1 M :X :x:1 I X I I 1 Ev

19721973

FIG. 8. The seasonal distribution, in number of individuals per day, of Aeneolamia

postica (Walker) (Cercopidae) from the wall lights of a house not far from the light trap.

The timing of the seasonal maxima

I have shown that many species do not occur throughout the year, at least not as active

adults, and of those that do, the vast majority has more or less well-defined seasonal peaks

in abundance. The question arises whether the seasonal peaks are in any way related to

the alternation of the wet and dry seasons. It is also of interest to see whether the seasonal

peaks of the different species all tend to occur during the same part of the year or whether

there is a tendency for the peaks to be spaced out, which would suggest that there are

strong interactions between some of the species.

In most cases it was not difficult to decide which week in the year was the peak week.

The maximum 4-week period used to calculate SM was taken and the week with the

highest number within that 4-week period is the peak used here. For several of the smaller

samples this did not work so well and then other criteria were applied, e.g. when there

was no one clear maximum and the sample was too small to decide about bimodality, the

meanpoint of the season was used. For species which, in my subjective judgment, were

clearly multimodal, all peaks were used in the analysis. The distribution of all the peaks

so obtained is given in Fig. 9 both for each year separately and for the 3 years combined.

There are very few peaks in abundance in the dry season. The large number of peaks

indicated in Fig. 9 in the early dry season, i.e. weeks 1 to 5, is mostly not real. That is,

they refer largely to species which lived in the lawns during the rainy season and now find

themselves without food in the dried out grassy areas and start flying around in large

numbers looking for suitable places to live. These maxima are peaks in activity rather

than peaks in abundance. In a number of cases I cannot be sure yet whether the suspected

peak in activity is that or is a real peak in abundance, so I have not weeded out the species

which have a maximum in the trap which does not correspond to a maximum in abund-

ance in the field. Part of the irregularities in the pictures for each year is due to the phases

of the moon. At full moon many species have a low abundance in the trap and thus fewer

species will have their maximum abundance during such weeks with full moon. The

general picture obtained from Fig. 9 is that the peaks tend to be spaced all over the rainy

season. Some of the peaks at the very beginning and some at the end of the rainy season

refer to seasonal dispersers, i.e. species that tend to have dispersal flights during those

times of the year and that do not necessarily live in the vicinity of the trap. The other



12/15

H WLDA287

20 a

I 0~~~~~~~~~~~~~~~~~~

20 b)

a

20o -Lz

ILi r

50 (d

40 -

30

20- IP

I0

01020304050

Week of maxima

FIG. 9. The distribution of the seasonal peaks over the years. See text for details. (a) 1974,

(b) 1975, (c) 1976, (d) 1974-1976.

species, then, may have a slight tendency to have their annual peaks during the middle of

the rainy season, but in general the peaks seem to be staggered rather than concentrated,

which suggest fairly strong interactions between the species involved, forcing them to

have their peaks spaced out. The dry season, apparently, is so inimical to most Homoptera

that they cannot have their seasonal peaks during that season, away from all the other

peaks of the rainy season. Only some species have been able to overcome this problem.

Geographical comparisons in seasonality

The measures SR and SM permit direct comparisons between different geographic

areas for which data are available. The ones discussed in the present paper from Las

Cumbres can be compared with data from Northern Finland (Kontkanen 1950), from

Kansas, USA (Blocker, Harvey & Launchbaugh 1972 and personal communication) and

from England (Waloff 1973).

The values for SR are plotted in Fig. 10. The one large value of SR = 50 for Kansas I

assume to be an error of estimation like the one discussed before, as the sample size for

that case was very small (n = 8). The differences between the four distributions are

highly significant as tested by x2 after grouping of the data. In Northern Finland the

length of the season for flying adult Homoptera is, on the average, very short, much

shorter than for most tropical species. In England and Kansas the seasons are of inter-

mediate length. Very long seasons, or species which occur throughout the year, are found

in the tropics only. However, short seasons, as short as the ones found in the subarctic,

are also found in the tropics of Las Cumbres, although there they are relatively rare. Data



13/15


I0 ? ab)

I n n u- AIol

10 (b)

n In1 rr, , i n n

10c

2o0 d 235

____ ___ ____ ___ ___ ____ ___ ___ 36-9

0102030405052

Seasonal Range

FIG. 10. The distribution of SR values in four samples of Homoptera ranging from the

subarctic of Northern Finland to Las Cumbres in tropical Panama. (a) Finland, (b)

England, (c) Kansas, (d) Las Cumbres.

from the salt marshes of North Carolina (Davis & Gray 1966) suggest that the seasons

of the Homoptera there may be as long as they are in the tropics (Wolda 1977) but the

data do not permit calculation of the SR values as sample sizes are not given.

The distribution of the species over the SM classes is given in Table 2. England and

Kansas are here not significantly different from each other, but all other differences are

highly significant. Sharp seasonal peaks predominate in Finland, but are also found in the

other areas, including the tropics. Although, on the average, the peaks in Las Cumbres

are much less pronounced than those of Finland, with Kansas and England occupying

intermediate positions. Unfortunately, the data given by Davis & Gray (1966) do not

permit calculations of the SM values for the Homoptera of the 'subtropics' of North

Carolina.

Annual Variability

Changes in abundance from year to year can best be discussed in terms of a parameter

called Annual Variability (AV, see Wolda 1978b). This is, briefly, the variance of the

distribution of log R, where R, indicates the change in abundance (Nt+1/Nt) for the ith

species. As a measure for abundance in a given year the total number of individuals

caught in the trap is used. It is usually advisable to include only those data where the

smallest of the two N-values is at least five (Wolda 1978b).

For the years 1974/75 AV for the Las Cumbres Homoptera is 0. 103 (N = 197) and for

1975/76 AV is 0. 115 (N = 201). These values are what one would expect for data from

humid areas, tropical or temperate (Wolda 1978b).

As I have discussed before (Wolda 1978b), one might expect species with a sharp

seasonal peak to be more variable in abundance from year to year, to give larger values of

AV, than species which do not have such sharp peaks, or do not have clear maxima at all.

Some data, including part of the data discussed here, supported that hypothesis. The Las

Cumbres data in their entirety are now re-analysed to test this hypothesis. Annual



14/15

H WLDA289

Variability is calculated for those species which have, in 2 successive years, values of

SM less than 4 and for those species which have, in 2 successive years, a value of SM

larger than 6.5, or, if the sample was small, larger than 7. Otherwise the effect of sample

size on SM is ignored here. The latter species have relatively sharp seasonal peaks, the

former species do not. The latter group has an AV of 0.176, the former has an AV of

0.075. The difference is highly significant, again supporting the hypothesis.

DISCUSSION

The seasonal distribution of insects, or of other organisms, is usually discussed with the

aid of long tables or long series of figures of various kinds. Several examples can be found

in Lieth (1974). Without some kind of a formal analysis of such data, however, the

interpretation of the results is rather difficult. To the best of my knowledge the present

paper is a first attempt at a quantitative comparison between insects from the subarctic,

the temperate zone and the tropics in terms of their seasonal distribution. Some of the

results are hardly surprising. One might have expected to see that many tropical insects

occur throughout the year as active adults, while subarctic insects have very short active

seasons. However, up to now that expectation was not much more than a hypothesis and

seasonality at different latitudes has not been properly documented. The present data,

to a large extent, conform to expectation. However, there are also some surprises. The

fact that seasons as short as those found in the subarctic also occur in the tropics is

unexpected. The dry season, inimical to most species, has a strong effect on the seasonal

distribution. However, species which are effectively non-seasonal, although uncommon,

do occur in Las Cumbres in spite of the existence of a strong dry season. Some species

manage to exploit the dry season conditions such that they have their annual maximum

during that season. A few even occur in the dry season exclusively and are absent during

the rainy months of the year.

An explanation of the various patterns of seasonality found in the tropics at this stage

would be premature. For the vast majority of the species dealt with in the present paper,

the life history is completely, or almost completely, unknown. In the absence of informa-

tion on foodplants of nymphs and adults, on predators, competitors, etc. I prefer to

postpone speculations, however easy these would be when there are no data to set limits

to one's imagination. The distribution of the seasonal peaks as given in Fig. 9 suggests

that interspecific interactions of some kind are an important factor in the determination

of the seasonal distribution of Homoptera in the tropics.

ACKNOWLEDGMENTS

It is a pleasure to express my gratitude to Miguel Estribi and Saturnino Martinez who

helped to sort the insects and to my wife Tineke who took care of the trap whenever I

was not there. The help received from Michel Boulard (Paris, France), Lois O'Brien

(Tallahassee, Florida), David Young (Raleigh, North Carolina) and Dwight DeLong

(Columbus, Ohio) in the identification of species is gratefully acknowledged. I am

particularly grateful to Derrick Blocker (Manhattan, Kansas) for sending me the only

partly published data for use in my comparisons.

REFERENCES

Blocker, H. D., Harvey, T. L. & Launchbaugh, J. L. (1972). Grassland leafhoppers. I. Leafhopper

populations of upland seeded pastures in Kansas. Annals of the Entomological Society of America,

65, 166-172.



15/15


Caballero, J. M. (1974). Meterologia, afno 1973. Estadistica Panamenla, 34, Serie L.

Caballero, J. M. (1975). Meteorologia, afio 1974. Estadistica Panamefna, 35, Serie L.

Caballero, J. M. (1976). Situaci6n fisica, meteorologia, afno 1975. Estadistica Panamenta, seccion 121,

clima.

Caballero, J. M. (1978). Situaci6n fisica, meteorologia, afio 1976. Estadistica Panamefna, seccion 121,

clima.

Davis, L. V. & Gray, I. E. (1966). Zonal and seasonal distribution of insects in North Carolina salt

marshes. Ecological Monographs, 36, 275-295.

Ehrlich, P. R. & Gilbert, L. E. (1973). Population structure and dynamics of the tropical butterfly

Heliconius ethilla. Biotropica, 5, 69-82.

Kontkanen, P. (1950). Quantitative and seasonal studies on the leafhopper fauna of the field stratum

on open areas in North Karelia. Annales Zoologici Societatis Zoologicae Botanicae Fennicae

'Vanamo', 13, 1-91.

Levins, R. (1968). Evolution in Changing Environments. Princeton University Press, Princeton.

Lieth, H. (1974). Phenology and Seasonality Modelling. Springer, New York, Heidelberg, Berlin.

Owen, D. F. & Chanter, D. 0. (1972). Species diversity and seasonal abundance in Charaxes butterflies

(Nymphalidae). Journal of Entomology, Series A, 46, 135-143.

Slobodchikoff, C. N. & Parrott, J. E. (1977). Seasonal diversity in aquatic insect communities in an all-

year stream system. Hydrobiologia, 52, 143-151.

Waloff, N. (1973). Dispersal by flight of leafhoppers (Auchenorrhyncha: Homoptera). Journal of

Applied Ecology, 10, 705-730.

Wolda, H. (1977). Fluctuations in abundance of some Homoptera in a neotropical forest. Geo-Eco-

Trop, 3, 229-257.

Wolda, H. (1978a). Seasonal fluctuations in rainfall, food and abundance of tropical insects. Journal of

Animal Ecology, 47, 369-381.

Wolda, H. (1978b). Fluctuations in abundance of tropical insects. The American Naturalist, 112,

1017-1045.

Wolda, H. (1979a). Abundance and diversity of Homoptera in the canopy of a tropical forest. Ecological

Entomology, 4, 181-190.

Wolda, H. (1979b). Seasonality parameters for insect populations. Researches on Population Ecology,

20, 247-256.

(Received 4 May 1979)

estacionalidad de insectos.pdf

Documents

Transcript of estacionalidad de insectos.pdf