Return to front page home
Return to spider life cyclespiderlink
Late Summer: How many molts?
MatingWith the amount a Nephila spider grows at each molt being fixed (not changing much with the amount of food or other environmental conditions; see Growth by molting page), the number of times the spider molts tells us both how old it is and how big it is when it matures.  And in Nephila, the number of times each individual molts is very dependent on whether it is male or female.
cartoon presentation of spider development
Male Nephila clavipes reared in the laboratory start out indistinguishable from their sisters.  They are the same size as babies, build webs, eat flies, and grow by molting. But after somewhere between 4 and 6 molts, they reach sexual maturity and never grow again.

In contrast, their sisters continue molting - up N. pilipes on my handto an additional five molts more.  This produces the giant female size that these spiders are known for. 

In this picture, a mature female N. pilipes rests on my hand - she weighed over 6 grams, more than many hummingbirds.
                  Human sexual size dimorphismSexual size dimorphism (SSD) is the phenomenon where males and females are drastically different in size.  As mammals, this is familiar to us in the pattern of males being larger than females. Because of the mammalian perspective of biologists, this was for a long time viewed as being "normal."

But when we look at animals that are not mammals, we find that in the vast majority of species, females are larger than males (among vertebrates, the most famous are the deep-sea angler fish, where males are reduced to minature external parasites of the females).

The Nephila spiders exhibit themale kleptoparasitizing most extreme sexual size dimorphism of any terrestrial animal.  Although males are not parastically attached to females, they are dependent upon them as kleptoparasites: males loose the ability to spin sticky silk when they mature, and must steal prey from a female. Only males that find a tolerant female will survive.

In this photo, the male on the left has stolen a prey item from the small juvenile female.
photo courtesy of M. Kuntner

For males and females to evolve to such different developmental pathways indicates that the benefits and costs of being small or large are very different.  We have a much better idea of why female spiders should be very large than why male spiders should be very small, so I'll talk about females first.

Size matters.
Darwin in 1871 proposed that large females are favored through a form of selection he termed "fecundity" selection:  larger females can lay more eggs.  In Nephila species I've studied, larger females lay many more eggs than smaller females - the figures below show my data for N. pilipes (left) and N. clavipes (right).  Bigger spiders (measured by the length of one of the leg segments) lay many more eggs than smaller spiders, supporting Darwin's hypothesis.  (That big spider on my hand is the upper-most point in the left-hand graph).
spider fecundity increases with size     fecundity increases with increasing spider size

If you look carefully at these figures, you can see that the X and Y axes are very different: the smallest N. pilipes (left) is bigger than most N. clavipes (right), and lays proportionaly more eggs.  Within each  species, the largest females are one or two molts larger than the smallest females.  Yet our best measure of evolutionary success – the number of eggs laid – increases when a female delays maturing by going through additional molts.  Why, if getting bigger is so advantageous, should any female mature at a small size?

The price of delay. In a nutshell the answer to this question is that there is an end to all good things, and in Mexico this is the end of the growing season.  Nephila clavipes populations live in habitats with definite changes in rainfall and temperature.  Spiders caught by the shift in seasons die.  How does this alter their size?  We can use a graphical model to predict changes in development.

growth model 1So if every spider grows the same amount each time it molts (the steps up), but the amount of food determines the time between molts (the horizontal lines), then spiders getting different amounts of food will mature at different times.  So long as the weather is nice, there is no problem.

growth model 2But if the spiders are living someplace where the weather is really different between the summer (growing) and winter (dormant) seasons, then the slowly-growing spiders might be in trouble.  We can draw a vertical red line to indicate the end of the season. The spider growing along the purple trajectory isn't mature yet when the season ends, and she dies.

growth model 3Her story has a happier ending if she has some flexibility in how many molts she goes through. If the spider growing along the purple line can anticipate the coming change in season, she can molt to maturity at the very next instar.  She'll be smaller than spiders who got more to eat and grew faster, but she has some chance of reproducing before the season ends.  Since having some offspring is better than none, she is more successful at this small size than if she had delayed maturing and died without reproducing.
I tested this model by looking at data I've collected over the years in different seasonal habitats.  There is a lot of variation in how well spiders do at capturing insects (mostly determined by their luck at choosing a good spot for their web).  If my hypothesis is right, then in seasonal environments, late-maturing female spiders will be smaller than early-maturing spiders.  I don't expect this to be the case for males, because they mature so early that they have no problem with growing too slowly.

Testing the model
.  When I looked at the size of females maturing at different times over the growing season in Mexico, the data clearly support the model of slowly-growing females making the best of a raw deal.  In all sites, late females are smaller than early females. Below are the data for females from all the Mexican sites and Panama (lower right panal).  Although the effects are slight, across all Mexican sites the late females are significantly smaller than earlier females. To reinforce the idea that the length of the season is important, the spiders are bigger in places with longer seasons (coastal Veracruz: Los Tuxtlas, Playa Escondida and Nanciyaga) and smaller in places with shorter seasons (Fortin, Tehuacan, Arroyo Frio, and Chamela).

This contrasts with the data from Panama, where there is no end to the growing season (although the wet and dry seasons are distinct) and there is no change over time in spider size. However, the wet season spiders capture more food, and they mature at a larger size. 

wet site femalesdata from dry sites
(from Higgins 2000 Oecologia 122:51-59)

So why are males tiny? As pointed out by Lande and Slatkin in the 1980s, males and females share a lot of genetic information.  Therefore, selection for enormous female size should (all else being equal) "pull" males along - the males should also be evolving to larger size.  Working with Jon Coddington, Matjaz Kuntner, and Charles Goodnight, I tested this by comparing male and female size across all species of Nephila.

spiderlink This link takes you to a test of whether, across all Nephila species, male and female size evolve in parallel. 

(Lande 1980: Evolution 34:292-307; Slatkin 1984: Evolution 38:622-630). 

spiderlink Although we tend to always think "big is better" (particularly if, like me, you spent a lot of time in Texas), there are costs to being huge. So another approach is to consider what factors might make large size disadvantageous.  I am only beginning to explore the negative consequences of large size, but this link takes you to my results thus far.   
Publications related to the material in this web page

L. Higgins. 2000 The interaction of season length and development time alters size at maturity.  Oecologia 122:51-59.
L. Higgins. 2002 Female gigantism in a New Guinea population of the spider Nephila maculata.  Oikos 99:377-385
L. Higgins and M. A. Rankin. 1996 Different pathways in arthropod post-embryonic development.  Evolution  50:573-582
L. Higgins. 1992 Developmental plasticity and fecundity in the orb-weaving spider Nephila clavipes.  Journal of Arachnology, 20:94-106