Why are dodders ectoparasitic?

Old pic of my C. subinclusa

Anyone with a passing familiarity with heterotrophic parasitic plants knows that they are mostly invisible. Monotropa, Rafflesia, Orobanche, Pholisma: all undetectable lumps and/or strings buried in their hosts or in the ground, shortlived flowers notwithstanding.






So why the hell doesn't Cuscuta do that?






The dodder literature is silent on the matter, but we can make a few deductions. First off: Cuscuta is in the bindweed family. This seems a bit too on the nose, if you ask me, like it's from some sort of video game with a questionable understanding of biology that went "oh they don't look similar but they're related because they're both twine vines!" What's next, dodders with wings?
I mean, monotropes don't superficially look like free-living members of Ericaceae either but it doesn't feel that weird to me that they're related.

But obviously this is real life so Cuscuta really is in the Convolvulaceae apparently [shrug]. According to a low-effort glance at the taxonomy literature the genus's position in the family is unclear, but I do think that the twining ability shared by dodders and morning glories is a synapomorphy of the clade (I have no proof but I do feel that this isn't one of the cases where climbing behavior convergently evolves multiple times within a family).

If we assume this to be true then the evolution of Cuscuta's particular growth habit is not particularly surprising. One is tempted to conclude that evolutionary constraints have so far prevented dodders from becoming full endoparasites, and that this is a highly imperfect solution because evolution is imperfect, and/or that perhaps given more time to evolve the distant descendants of dodders may circumvent said constraints and become fully internal parasites.




Is that really the story, though? I mean, don't get me wrong, evolution is imperfect but I find that it is not as imperfect as people tend to think sometimes. A lot of traits that get handwaved away as "oh just arbitrary stuff caused by phylogenetic history, not everything the blind watchmaker does has to have a function" actually turn out to be pretty important adaptive features.

I suspect external climbing is an important part of the dodder life history strategy. The dodders I've worked with (and seemingly other dodders, from what I've seen online) appear to be tender-shoot specialists, and apparently die of malnutrition on sufficiently mature stems even when they manage to penetrate them. Which is a bit weird, since mature stems carry nutrients to the more nutritious juvenile tissues at their tips and if you've encircled the former it seems like you could intercept the nutrients going to the latter right? But leafhoppers and aphids able to feed on mature stems also tend to preferentially move to and feed directly on tender shoots when they can afford it, so I suppose whatever logic governs nutrient extraction efficiency applies to both the herbivorous insects and herbivorous plant in question. Also, dodders seem to feed only or mainly on plant parts they are wrapped around and not the rest of the plant; I have repeatedly seen strawberry petioles and nightshade stems that have been wrapped rapidly wither and die while all adjacent uninfested ones on the plant remain healthy. Though I have seen both C. subinclusa and C. californica form green islands on dying host tissue in certain circumstances (though not others), so some limited level of nutrient redirection seems likely.

Suppose you are a dodder and wish to maximize your intake from tender shoots while minimizing the volume of tissue you grow, given realistic environmental constraints. To get from one tender shoot to another on an adjacent branch, an obligate stem endoparasite would have to grow all the way down the branch and all the way up the other branch. Why do that when you can just bridge the gap by growing across? It's much less of a distance, and air is presumably a less resistant medium to grow through than rigid plant tissue. Plus being ectoparasitic allows dodders to span multiple hosts at once; this seems not that common in wild adult subinclusa and californica, but is apparently a major means of spread for temperate dodders, and juvenile subinclusa (in order to get at the part of the tree that's nonwoody enough to actually penetrate mechanically, since regardless of how nutritious it is or isn't the main trunk is just too thick to drill through) often start out feeding on small herbaceous plants and gradually climb taller and taller plants until they reach the edible portions of the tree.

You may have seen me occasionally complaining about my dodders mindlessly climbing upwards the same way ladybugs and a lot of other foliage-inhabiting insects do; while it's definitely pretty annoying in captivity (since when the old host is exhausted and you put a new host that's shorter next to it the dodder often struggles to find it), the adaptive benefits of this in nature are pretty clear.