Growing sand-verbenas (Abronia/Tripterocalyx)

For awhile, I've been working on plant defense and other aspects of the ecology of the sand-verbenas (here and here). Despite being much-beloved coastal and desert wildflowers, Abronia species have a (slightly deserved) reputation for being hard to germinate, hard to keep alive, and just generally not that great plants. Thus, they are uncommon in gardens of even the most diehard native plant folks.

Abronia villosa, Imperial Dunes, 3/2017. 

For the past two years, I've been growing them in the lab, greenhouse, and home garden  and refining my protocols for growing them. I haven't taken a really scientific approach to this - I haven't done huge numbers of replicates or really controlled environments, but I've tried out a number of approaches, talked to lots of more knowledgeable folks, and tried quite a number of species with success.

Abronia maritima, the easiest species to clone and perhaps the easiest to grow in captivity. Morro Bay, 4/2016. 

The keys to success are actually quite simple - (1) they need sand, (2) they need space, (3) if you want flowers, they need fertilizer.

(1) Most sand-verbenas are restricted to very sandy areas - coastal dunes, desert dunes, desert washes, etc. so its definitely no surprise they need sand. For bigger plants, I use ~75% play sand - the cheapest at ACE - and ~25% potting soil (I've been using Sunmix with good success). I've tried 50/50 and some species (such as fragrans), do like this, but the sandier coastal and desert ones seem to like more sand. Sand is really well-draining, so you needn't worry about overwatering, but it does lose its moisture really fast, especially if the plant is too large for the pot, so keep an eye on them. I water every 3-4 days in the lab and greenhouse.

For seedlings, I use closer to 50/50 and keep them covered to retain moisture. They dry out very quickly and it takes them awhile to really get going (much longer than most of your garden plants).

I tried transplanting some into the clay soils of Davis. Unsurprisingly, they all died very quickly. I bet you could do them in turface, cactus mix or perlite or maybe even coarse vermiculite, but sand is proven and it is so cheap and easy, I don't feel much need to change that.

Abronia pogonantha in the lab. You can see the flowerbuds just starting on this one. 

(2) Keep them in big pots. This is really important - Abronia usually have really extensive root systems to find whatever little nutrients and water are in their sandy environments. The picture below demonstrates this very clearly.

The root system of Abronia latifolia in a blown-out dune at Doran Beach, Sonoma, CA. 

The desert species - villosa, pogonantha, turbinata - can get away with being in small pots - but if so, do use the really deep ones. However, since they are all sprawling, prostrate groundcover, they look and do best if given space above-ground, too. I transplant them to 8" or 10" pots when they have 2-4 leaves. This allows them to develop a good root system fairly quickly and seems to lead to quicker growth and flowering. On my patio at home, I have villosa and fragrans in 5 gallon buckets (with holes drilled in the bottom) and a ~10 gallon planter. Both of these have ~75/25 sand and work well (and buckets are cheap and easy).

Abronia ameliae in a 4" pot. This plant desperately needs transplanting. 

(3) Fertilize constantly. Sand (and the nonnutritive potting soil I generally use) is pretty much nutrient-free. This seems to not matter much for vegetative growth, but for most of the species, I get flowers only when I fertilize often (1/week or more!). The annual species seem to flower with a bit less pushing, but the maritime species, especially umbellata and maritima, need to be given a little push. With regular fertilization, warm temperatures and longer days (15 hours or so), they will flower constantly - I have fragrans, umbellata, and villosa that have flowered for over a year without stopping.

The long-lived and long-flowering umbellata individual. Cloned from field-collected tissue in SLO county. 

Germination. I'd written a post before on germination, but I won't even link to it, as my germination protocol has gotten much better. Here's how I do it now. You'll need ethephon (I use Monterey Florel Brand Growth Regulator, available online or at Home Depot) and some sterile media - I use vermiculite, but paper towels work, too and tupperware or a petri dish.

Papery fruit (=anthocarps) of Abronia latifolia. 

(1) Remove seeds from the papery fruit. Note that many of the fruit may be empty or have inviable seeds. I'm not sure what causes this, but its likely pollination failure.

(2) Mix a solution of ethephon. For the Monterey hormone, I use .66mL per 1L of water (distilled is preferred).

(3) Place seeds on vermiculite or wet paper towel in your container, moisten heavily (but don't wet - there shouldn't be standing water) with your mixture. Leave for a few days in a cool dark place. If you get lucky, it will look like this after a couple days. Leave them for a day or two more, until you see the roots starting to get fuzzy and elongating (>1 cm, at least).

Germinating Tripterocalyx micrantha seeds. 

(4) Transplant into a 50/50 sand/sterile soil mixture, well-moistened (but not wet), which is covered to keep the humidity up. They will grow painfully slowly. But, eventually they will get a hold on life and put on a few leaves. At that point, transplant to a larger pot and bear in mind the tips above.

Seedlings (of hybrid turbinata x pogonantha). Notice that they have one seed leaf - these are derived dicots, not monocots as it might be tempting to assume!

Growing from cuttings.

I've had alright success with cloning plants and some folks at the botanical conservatory here are doing a little more systematic attempt at figuring out how to clone them. I was using low-strength rooting hormone (1%), cutting below a node, and putting them in sand/perlite. I was getting low success, but importantly, not no success. A. maritima and fragrans were the species most amenable to this - I've had mixed luck with others (turbinata/umbellata/ breviflora) and complete failure with latifolia and villosa. The successful ones took a really long time to root - 4-6 weeks - but once rooted were quite happy and as easy to care for as others.

Hopefully I'll have an update on this soon.

A. umbellata umbellata (left) and A. latifolia (right). Morro Bay, 5/2017.

Other various tips:

I cage my plants with chicken wire to keep them somewhat contained. Tillett (1967) says that he trained his up poles in a greenhouse. I've had no luck with this, but the chicken wire cages work pretty well. I either route the stems coming out back in, or just trim them off. This seems to work well to conserve space.

From left to right: A. umbellata breviflora (pink), maritima (behind, upright, no flowers), Tripterocalyx micrantha (in front, drooping), latifolia (behind, dark green leaves), turbinata (white flowers), umbellata umbellata (purple, far right)

Crossing is pretty easy. Since most species (A. u. breviflora, all Tripterocalyx spp., and A. ammophila excepted) are self-incompatible, you don't generally need to emasculate. Find out where the stigma comes up in the tube, and cut the flower between that and the lowest anther, open up the tub so the stigma is exposed, and pollinate! I've had good luck with intra- and inter-specific crosses.

Hybrid pogonantha x maritima. 

If you have any comments or questions, do let me know! I'd be really excited to help folks out who want to try them. 

Data I'll never publish II: Salinity and herbivory

I spent a lot of my second year of grad school thinking about salinity and insect herbivory. Generally, insects don't like very much salt (i.e. how many marine insects have you seen?). Salt is a fairly effective herbivore deterrent - an observation seemingly first made in 1980 by D. Newbery in an Oecologia paper on mangrove herbivory. I made the same observation, and tested it experimentally, in chenopods in a 2014 paper (also in Oecologia - they've seemingly cornered the salinity/insect herbivory market).

Coconut palms might be the most widespread and useful (to human) halophytic plant. They were useful for that hammock, at least.  Abaco Island, Bahamas, 2011. 

Plants are also affected by salt and have myriad ways to deal with it, basically all variations on either excluding it, sequestering it, or excreting it. Obviously some plants are much better at dealing with salt than others (see mangroves, Zostera, etc.) - we call plants that are adapted to saline environments "halophytes" (i.e. salt plant in Greek). I happened upon a little, weedy, nonnative, and pretty much unremarkable chenopod - Oxybasis glauca - growing at the edge of a building in Davis and somehow I decided it was a pretty cool plant. Given all the other cool halophytes available, I'm not sure why I chose this plant to do a bunch of experiments on, but I did.

This is Oxybasis glauca growing in volcanic sand on the edge of Mono Lake, Mono, CA. I was with a group of about 30 people when I found this and was very excited. I couldn't really even articulate a single cool thing about the plant - it is salt tolerant, but every plant in that area is salt tolerant. Maybe the coolest thing is that Oxybasis species have really small seeds compared to Chenopodium or Atriplex... maybe there is nothing special about it?

Like most Atriplex and Chenopodium (the genus which Oxybasis was split from) species, Oxybasis glauca has salt bladders - little bubble like trichomes which the plant shunts salt to and then they burst, an odd but effective form of salt excretion. This leaves a layer of salt on the outside of the plant. This protects the plant from herbivory somewhat.

Pre- (above) and post- (below) bladder burst O. glauca leaves (lab-grown). 

Because O. glauca is salinity-tolerant and the primary herbivore of most weedy chenopods in the valley, the spotted cucumber beetle (Diabrotica undecimpunctata), doesn't like salt (see my 2014 paper), I wondered if there might be a refuge from herbivory effect at higher salinities and maybe there would be an intermediate salinity where the plant would still grow well, but herbivores would be deterred. So I did an experiment - I grew plants in three salinities* and then exposed half of them to a week* of cucumber beetle herbivory. I expected herbivore pressure would be most intense at low salinities, but also growth would be retarded at higher salinities. 

So the hypothesis looks something like this - if plant "performance" is on the y-axis and the green line is effect of herbivory and grey the effect with solely salinity, if there is some overlap, the plant might do best at that overlap point (or it might not). (note: this is not a particularly good graphical representation for a number of reasons). 

What did I find?

Plant response to salinity (w/o herbivores):

Salinity increasing left-right. Standard deviation plotted.

Plants did worse as salinity increased (as expected). 

Herbivory:

Salinities increasing in treatments 1-4. Standard deviation plotted.

Total leaves damaged by the herbivores decreased with increasing salinity (as expected, as they are less palatable), but because the plants had fewer leaves, the proportion damaged increased. 

THE INTERACTION

Biomass of plants. Dark green: with herbivores, light green: without herbivores. Salinity increasing left to right. Standard deviation plotted. 

Sadly, there wasn't. Beetles didn't really have an effect on biomass (or any other metric). Maybe I didn't have them in there for long enough? Maybe they really don't have a fitness effect (I can certainly believe this). 

Maybe this data will be useful to someone. Email me for the sheets. 

*Note: the exact procedures are in one of about 40 notebooks in my office, so I don't actually know exactly the salinities or number of days right now. If anyone is interested for any reason, I can easily dig this up.    

Musings on pollinators and sticky plants

Having spent much of the last four years seeking out sticky plants, I'll admit to mostly having thought about their herbivores and general insect communities on the plant surfaces, without too much thought to pollination. A labmate's question about columbine (Aquilegia eximia) pollination - "Do they have any adaptations to prevent bees?" got me thinking about this. A. eximia is hummingbird-pollinated and is extremely sticky (see here). I've spent enough time looking at, and photographing, over a hundred species of sticky plants that I was able to dredge up some pertinent observations (some from field notebooks) and I'll reference a few papers that have mentioned this.

A honeybee entrapped on the sticky flower/pedicel of serpentine columbine (Ranunculaceae: Aquilegia eximia). Note the robbing hole present on the front (out of focus) and left spur. These were made by carpenter bees (Xylocopa californica). Honeybees and other bees visit as secondary robbers, though they do not regularly visit unrobbed flowers. The primary pollinator of A. eximia here is Anna's Hummingbird (Calypte anna).  McLaughlin Reserve, Lake County, CA. 

The general explanation for sticky plants is that it prevents herbivory directly (by slowing or entrapping herbivores). This is generally an insufficient explanation, as pointed out by Tom Eisner and colleagues (here) who found that ladybugs were slowed or entrapped, allowing a specialist aphid to feed on Mentzelia pumila (Loasaceae - a painful plant family). Eisner's explanation, however, also seems simplistic and incomplete - it turns out that a suite of specialist predators are good at moving on sticky plants and do effectively prevent herbivory (here, here, here, more coming!). Whether these are exceptions or the rule, I suspect this defense is fairly common. While I think Eisner et al. missed the boat on trophic relations in Mentzelia (this is not a completely baseless accusation; I have some data to back my case up), they do make an astute observation on a cost of being sticky: "among the insects we found dead on M. pumila there were several individuals of an andrenid bee, Perdita sp., ... a genus known to include pollinators of Mentzelia. Evidently the trichomes can be a hazard to [pollinators] ".

People have long thought about adaptations both for a specific pollinator and against others that are not as effective. Is stickiness a problem for pollination or could it be a benefit in some cases?

Entrapped Lycaenid butterfly (I think it is Plebejus acmon) on Mentzelia micrantha. It seems likely this butterfly was entrapped while visiting a flower, as it is entrapped on a flowerhead (it is on a bract surrounding a fruit, but it probably had been there since there was a flower). Napa County, CA.

First, we should ask: do sticky plants actually entrap their pollinators commonly? Unlike the smaller-flowered Mentzelia (like micrantha pictured above), which are likely self-pollinating, the larger-flowered species, including Eisner's pumila (and the widespread and stunning laevicaulis) have large flowers separated from the very sticky (like velcro) leaves. Carnivorous plants, including the sticky sundews, often have widely separated flowers and traps. Yet, a nice study by Jurgens et al (here) found that the separation in one sundew is simply because taller flowers attract more pollinators (and the sundew's sticky leaves remain near the ground). They found that the color differences between the flower and trap were more important in keeping pollinators on the flower and off the leaves. Complementing this, a more recent study showed that color is important, but odor might also be important, especially in sundew species without much physical separation (here). A strange paper (see footnote) on a South American aster showed practically no overlap between pollinators and entrapped insects (here). The sticky columbines - eximia, shockleyi, and some populations of formosa - are all hummingbird-pollinated, as is the sticky monkeyflower, Mimulus cardinalis; the chance a hummingbird becomes entrapped is virtually zero. 

The first thing to note is that none of these papers have actually examined whether there is a pollination cost to being sticky. Of course, a past cost may shape evolution, but not be obvious now. While these examples make the case for putative adaptations to not entrap pollinators, none except El-Sayed's paper have any sort of comparative aspect. Therefore, we can't be sure that these are actually adaptations. This isn't to belittle the research, of course, as the experimental studies are very nice and convincing that these factors underly the non-trapping of pollinators.

A small bee of some sort entrapped on a serpentine columbine bud. Two Tupiocoris californicus, one of the mutualistic predators, are pictured on the right of the bud. McLaughlin Reserve, Lake County, CA. 

I wonder, if perhaps in addition to these chemical and visual cues keeping pollinators away, if another way to avoid entrapping pollinators is to utilize pollinators which are too large to be entrapped? Could it then be used to exclude suboptimal pollinators? Might stickiness might be easier to evolve in plants with larger pollinators or sticky plants be selected on to have larger pollinators? This chicken-or-egg situation need not be resolved; it would be interesting enough to find out whether there is a correlation. One could go about testing this by looking at insect-entrapping plants and comparing them to their close relatives and asking (quantitatively), is the pollinator of this sticky plant larger than expected given the pollinators of its closest relatives? Our paper on columbines includes a genus-level list of insect entrapping plants, this would be a good place to start when looking for possible comparisons (at whatever level you choose, you'd need to have both sticky and non-sticky members of each taxa). Off the top of my head - Aquilegia, Mimulus,  Salvia, Calceolaria, Ribes, Nicotiana seem like genera with both sticky and nonsticky members and pollinator variation and would be a good place to start (surely I am missing some).

A small solitary bee (I'm not good on bee genera) on Mimulus bolanderi. Among Mimulus that I've investigated, bolanderi entraps the most insects. There is little information on this strange, fire-following species, and I have only seen one pollinator visit (this one!), but I was mostly looking for predators. Another insect-entrapping Mimulus is hummingbird-pollinated (cardinalis). Some populations of common monkeyflower (M. guttatus) are sticky, others are not - all are bumblebee pollinated, I believe.

In conclusion:

1) Sticky plants occasionally entrap potential pollinators, though we don't know yet whether there is a realized cost to this, indeed it could be a benefit and may keep less-effective pollinators away.

2) Several experiments have demonstrated that features of some sticky plants prevent pollinators from being entrapped.

3) I hypothesize that sticky plants may have larger pollinators, as it would reduce chance of a cost (but I am agnostic about which might evolve first) and might even have a benefit (as an adaptation to exclude some pollinators).

I'd love other people's thoughts or other references (I haven't really dipped my toes into the literature on keeping less effective pollinators out of flowers).




Footnote:
I hope this group follows up on this system, as cool questions could be asked, but this paper is confusing (and if they read this post, they should read Patricia Thomas's 1988 dissertation at U. Illinois, which looks at a similar system in a thistle - I have a pdf). They looked at Haplopappus flowers, which have sticky bracts (like Grindelia and some Cirsium) and recorded flower visitors and entrapped insects to see whether there was overlap. This, in and of itself, is a good exercise. However, the paper is lacking in much pertinent information and clear discussion of what they found, What stuck out to me most was that they found that the tephritid fruit fly, Dioxyna chilensis, (which is an herbivore of this plant) was a common flower visitor and commonly entrapped, thus the stickiness may be a direct defense. Somehow this and really any herbivory isn't mentioned - despite "herbivores" being mentioned in the title - herbivores weren't really discussed in the paper. Also, they discuss pollinators, but not really any functional explanations of why some insects might end up on bracts and others on flowers (for instance, they found ants and a family of beetles were the most commonly entrapped insects - yet they don't discuss that both primarily walk, whereas the non-entrapped taxa fly). Their conclusions are technically right, but rather incomplete: "Thus, comparing all arthropod genera found, it seems that bracteal resin selectively traps insect genera with lesser pollen transfer potential." Hopefully more work will come out, as it seems Villagra's lab is working more on the system. 

An unexpected herbivore

Columbines are toxic! Like larkspurs, columbines are supposedly toxic to most livestock and humans. So say the books. This rabbit doesn't listen to the books. (The internet, in its infinite wisdom, says that the eastern species has edible - to human - flowers, so maybe the rabbit is just rather tech-savvy)

I actually suspect that the roots and leaves may be somewhat toxic but the reproductive parts, including the flowers and pedicels (which the same rabbit eats in the next video!), are not. Deer also eat them, especially in one particular population, which I've mostly stopped using for experiments because of it. Wild speculation aside, just thought I'd share the video as I got a kick out of it.

Also, a question - is this a brush rabbit (Sylvilagus bachmani), a European rabbit (Oryctolagus cuniculus) or a black-tailed jackrabbit (Lepus californicus)? . A terrible still of the tail from the video is below. I feel a bit silly that I can't even conclusively get it to genus. I'd be kicking myself pretty seriously if I couldn't get a dragonfly, bird, wildflowers, or butterfly to genus and really this should be far easier!

Another sticky columbine with dead bugs and predators!

A few weeks back, Rick Karban and I headed to the east side of the Sierras and the White Mountains to look for some Abronia and Nicotiana species for continuing projects. I also took a wish list of species (with herbarium GPS coordinates) that seemed worth checking out. We were able to easily find Abronia nana (not really sand-catching), Abronia turbinata (somewhat sand-catching), Nicotiana attenuata (bug-catching), and Alliciella monoensis (sand-catching). I also had a record of Aquilegia shockleyi, the desert columbine. After spending so many hours over a few years looking at eximia (here, here), I've spent more time looking at other columbines. The sierra species, formosa, is fairly boring from an insect-plant perspective, being nonsticky, largely devoid of predatory bugs and often covered in aphids.

This Aquilegia formosa was growing along Mormon Emigrant Trail (El Dorado county), seen on the way to the east side. Compared to my eximia, it has tiny flowers (probably 2/3 the size of eximia). With only anecdotal evidence to support this, I suspect that this is because Calliope Hummingbirds, the smallest one around, are a common visitor (though others are, too) and would be unlikely to successfully access the nectar in a much bigger flower.  

Aphid exuviae on a bud of A. formosa.

The Jepson said that shockleyi hybridizes with formosa in desert mountains, so I was expecting it to be fairly similar - nonsticky and not that interesting from my plant-insect perspective. Fortunately, I was wrong.

Aquilegia shockleyi, White Mountains, CA.

In a really pretty little desert oasis, we found a nice population of shockleyi growing in a flowing spring with my favorite thistle, Cirsium douglasii, an associate of A. eximia at my field site. This species is every bit as sticky as eximia, every stem had dozens to hundreds of entrapped insects.

A typical shockleyi stem - dead beetles, flies, wasps - tasty morsels for the right predator!

Another, for good measure.

Of course, my first thought was - is a carrion-mediated defense system happening in this species, too? That requires that there are scavenging predators and herbivores that the predators can eat (often they eat just the herbivore eggs). Sure enough, in just a few stems, I quickly found both.

The herbivore was, predictably, Heliothis phloxiphaga - a really common and destructive caterpillar that is common on most glandular plants in California and is multivoltine (has several generations per year). It is extremely bad for plants as, unlike many caterpillars, it largely eschews leaves in favor of buds, flowers and fruit (it is part of the aptly named group, the budworms). This is the same species that eats eximia and is controlled by the sticky plant predators at McLaughlin, where they consume its eggs and small caterpillars. 

A late-instar caterpillar on shockleyi. They have no trouble at all moving on the sticky stems of any plant, though I don't know exactly how they do it.

Typical Heliothis damage to a shockleyi fruit. It has consumed 3 of the carpels (the seed pods) completely and consumed all the immature seeds from the two remaining ones seen in the picture.

Predators were present as well. They were not in abundance, but were around. The most common was Hoplinus eschinatus - usually the most common on eximia, as well as tarweeds, sticky Mimulus, sticky tobaccos, etc. They are great egg predators and are just great insects in general. I got only crummy pictures that are not worth sharing here. The other interesting predator present was Oecantha - tree crickets. I only saw two really tiny nymphs (pictured below), but they are also on eximia in low numbers. Unlike most Orthoptera (the grasshopper/cricket/katydid order), they are really omnivorous, feeding on dead insects, live insects, as well as the plant in small quantities. On tarweeds, especially Madia, where they are common, they have benefits and costs the plants - they are predators and remove some herbivores, but they also chew leaves and oviposit into the stem, leaving big scars. I've never seen these scars or any evidence of them chewing on columbines (which are notoriously toxic to many animals), so I suspect that is less of a problem, but I really don't know. I've only seen them three or four times in three years of working on eximia, so I'd bet they are too rare to really have a huge impact - and these two were the only which I saw on shockleyi. 

Two tree cricket nymphs. 

There was one interesting contrast to the eximia system. A seed bug of some sort (I didn't collect it) was present and feeding on the seeds. I've noticed occasional clutches of eggs on eximia, but the nymphs either disperse or perish in the stickiness; I've never seen one actively feeding. This bug moved with seeming ease and was common. I'd bet that Hoplinus and other sticky plant predators would be effective at controlling it; the egg masses are present on the plant and the nymphs likely stick around on the same plant (somewhat important for the carrion-provisioning system to work). 

Unknown bug feeding on shockleyi seeds in a ripe fruit.

Just hatched nymphs of the unknown bug on a ripe shockleyi fruit.

In conclusion, this was a really cool plant, and it is a strong, strong, candidate for the carrion-provisioning system which I described in eximia. Someone should study it. A simple carrion-removal experiment, checked weekly through the growing season would be an easy project (for an undergrad, perhaps?) and would give us new insights (how do these sticky plant predators control seed bugs?) and would be really fun. There is nothing as much fun as being in a wet place in the desert or coast range, with hummingbirds visiting the columbines around you and other birds, mammals and insects abundant in the little watery oasis. 

Another picture of dead stuff on shockleyi stems. Please study it.