Thursday, June 19, 2014
Monday, June 16, 2014
So this is the one that was going to be about Anthurium genetics, but I've realized that even though I've read a bunch in the book and elsewhere, fairly carefully even, I still can't entirely grasp what's going on. It doesn't help that the book is almost 20 years out of date, and that even more complicated explanations are sitting out there on the internet, taunting me with even more details I don't understand, but I suspect the main problem is that I don't have any actual training in genetics, so everything I understand about how genes work is self-taught. That works out okay when it comes to concepts, but I get lost quickly when the specialized vocabulary gets going.
This is uncomfortable. It's especially uncomfortable because I understand enough of the individual pieces that I feel like I ought to be able to understand the whole. And also because understanding what was going on genetically was the whole reason I bought the book1 in the first place.
Specifically, I wanted to understand #097, "Colin Ambulance," to whom you were introduced in Part 1. Here's Colin again, to refresh your memory:
And here is Colin's seed parent, the NOID purple:
Even taking into account that I don't know who the pollen parent was, it would be hard for me to find a more opposite-colored mother/daughter pair. Aside from 'White Gemini,' I don't even have any varieties that are this light, and 'White Gemini' is probably not the pollen parent.2 But then, most of the NOID purple's children don't look much like her. Here are the six known offspring to date:
Now, Mario looks enough like his mother that it'd be easy to guess that they were related even if you didn't know. And although Aurora and Deena don't look much like their mother on first glance, that slight purple tint to the spadices would give away the relationship if you stared at them long enough. Maybe even Delta. But it's been a mystery to me how Colin and Mason happened.
Not to spoil the ending, but -- it turns out that that's just how it goes with Anthuriums, a lot of the time. The children don't necessarily resemble their parents very much. Not only do Anthuriums have multiple pigment-making genes, but the expression of those genes is moderated by other genes, one of the pigment genes makes the starting material that the other pigment genes work on,3 and some pigments mask the appearance of other pigments. Put all those genes and their interrelationships into a cross and mix themselves up, and you can cross spathe color A with spathe color Z, expecting to get spathe colors in between A and Z, like C, L, and W, and instead wind up with spathe colors ü, ¿, and ♂. (Though sometimes you also end up getting F, F, and more F -- even the unpredictability is unpredictable.)
The pigments involved in Anthurium spathes are pelargonidin (orange), cyanidin (red), peonidin (purple),4 and chlorophyll (green). These colors may be produced in varying quantities, depending on the genes that determine how much pigment to make. So cyanidin is responsible for red spathes, as you'd expect of a red pigment, but it's also (usually) the pigment in pink spathes, just in a lower concentration. Same thing goes for pelargonidin, which can be coral5 or orange; chlorophyll, which can be varying intensities of green; and peonidin, which changes color depending on the pH where it is, and how much of it there is, but runs from pink to purple.6 The varying intensities of color, in varying combinations, give us all the different possible colors of spathes.
Pelargonidin, cyanidin, and peonidin are all either being made or not being made, under the direction of one gene for each, and then another gene influences how much is made. (It looks like each pigment has at least one corresponding intensity gene, though I may be misunderstanding that.) But the point is that when you shuffle the pigments around during pollination, sometimes the intensity genes get shuffled around too, and a red crossed with a red may yield a pink, for example.
And the genes influence one another, besides. The gene for "make red pigment" I think uses orange pigment as the starting material; that is, it produces orange pigment, then turns some of it into red pigment. So unless I've misunderstood something badly,8 you can't have a red spathe unless there are genes present for making both orange and red. Add to this the possibility that plants with peonidin in them may have different pH in the spathes, and consequently may be producing different colors out of the same genes, and the tendency of red pigment to mask the presence of orange or green pigment, and you can see how variable things might get. The authors of the book at one point crossed two pinks together and wound up with, by their count: 5 bright red, 19 red, 42 light red, 4 dark pink,9 7 pink, 2 orange, 21 coral, 10 light coral, 1 white with a touch of pink,10 and 29 white.
So you can see how complicated and frustrating being an Anthurium breeder could be, even before you factor in the consequences of randomly pollinating everything with whatever's handy.
Since I haven't been able to do much deliberate crossing, due to the fact that Anthurium inflorescences don't spend that much of their lifetimes shedding pollen or accepting pollen, and the fact that up until pretty recently I didn't have that many plants blooming at once, the semi-unpredictability isn't that big of a problem. It's tough to be frustrated by things standing in the way of your goals when you aren't working toward any particular goals. (In fact, so far, it's been plenty exciting to find out that I can make seedlings at all -- none of the houseplant books talk about growing Anthurium from seed.11) I've basically been a toddler mixing together all the finger-paints and seeing what happens. New colors! New shapes! Cool!
And I'm likely to continue finger-painting for a while: I just don't have the facilities to determine what genes I'm working with. In the book and the paper both, the procedure for figuring out what they had was that they just made deliberate crosses between two plants, however many times were necessary to get enough seedlings to be statistically relevant, and then they just grew them all out and counted the colors at the end, for as many different varieties as necessary. Which, we're talking about 75-100 seedlings for each cross, and doing a single pair of plants doesn't tell you much about either plant's genes: you have to compare what you get when you cross each of those two plants with a bunch of others. With 13 cultivars in my founding population here,12 that's 91 different possible combinations, with each combination producing, say, 75 seedlings, and at minimum, I'm stuck growing out almost 7000 seedlings for two years just to be able to make reasonable guesses about the genetics of what I have. In the meantime, new seedlings have arisen that I will probably also be interested in using for future breeding, so I'd need to cross some of those out as well, to determine what genes they wound up with. So breeding everything out until I understand what genes are present doesn't work. And genetic sequencing is way too expensive, plus I don't think the relevant genes have even been identified, so having the sequences wouldn't necessarily do me any good even if sequencing were affordable. I'm just doomed never to understand exactly what's going on.
So, you know. Fuck it. I might start getting a little more careful with the crosses, and aim in certain directions (In particular, I would love to create a decent-sized, blistered, dark purple spathe, or any green spathe at all,13 especially if the leaves were nice, too.14), but we're probably looking at more finger-painting. Which is fine. If I've learned anything from all this, it's that it's relatively easy to get new colors; the hard part is getting new colors on purpose.
(► indicates seedlings making their first photo appearance on the blog in the mid-June posts)
First column, top to bottom: 005 ("Chad Michaels"), 066 ("Barbara Seville"), 108 ("Deena Sequins"), ►125 ("Anya Wei"), ►232 ("Rhoda Badcek"), ►231 ("Rhea Listick"), 245 ("Sawyer Ad"), 235 ("Rowan DeBoate").
Second column: ►097 ("Colin Ambulance"), 059 ("Bijoux Tuit"), 126 ("Erin Dirtylondry"), 031 ("Sylvester"), 223 ("Patty Cake""), ►244 ("Sara Problem"), 116 ("Eileen Dover"), 282 ("Dave Trading").
Third column: 046 ("Aurora Boreanaz"), 035 ("Alyssa Edwards"), 076 ("Bob Humbug"), 200 ("Mario Speedwagon"), 276 ("Zach Religious"), 243 ("Sal Monella"), 271 ("Wanda Reulthemal"), 234 ("Ross Koz").
Fourth column: 026 ("Peaches Christ"), ►283 ("Anne Pursand"), 239 ("Russ Teanale"), 149 ("Heather Boah"), ►216 ("Gillian Jamm"), 280 ("Jujubee"), ►247 ("Selma Carr"), 273 ("Wes Coast").
Fifth column: ►083 ("Carmen Adairya"), 063 ("Audrey Quest"), ►202 ("Mason Pepperspray"), 085 ("Carson Trucks"), 238 ("Rudy Day"), ►110 ("Delta Badhand"), 118 ("Elijah Sturdabowtit"), 275 ("Yvette Horizon").
Coming up in Part 3: what sorts of things do Anthurium breeders care about, and why? (This will probably take me several days to do, since there are ongoing family things happening, and it's a complicated subject besides. I hope to get a product review post in before Part 3 as well, but we'll see what I'm capable of.)
2 'White Gemini' probably isn't the pollen parent because I can't remember ever seeing 'White Gemini' produce pollen. Anthurium spadices accept pollen before they make it, so any spadix that's highly fertile, like 'White Gemini,' gets pollinated before it makes any pollen of its own. It's also the case that some varieties just never make pollen, though I don't have any reason to think 'White Gemini' is one of those.
3 Such that if both copies of the "make orange pigment" gene are broken, the plant will produce white spathes, even if there are also genes telling the plant to make lots of red pigment. Without the ability to make orange, there's nothing for the plant to make red out of.
4 Actually those three are all the 3-rutinosides, but the pelargonidin/cyanidin/peonidin parts are where the color's coming from so it's easier to use the short, inaccurate form of the molecules' names. I know, like you even care.
5 Anthurium geneticists use "coral" to mean light orange, though I've always thought of it more as the medium-brightness color midway between pink and orange. Google image search for "coral color" suggests that as a color, "coral" can mean pretty much anything you want it to mean, as long as it's not dark and resides somewhere in the pink/red/orange neighborhood. Clearly someone needs to set some guidelines, but in the meantime, we'll go along with the scientists' fucked-up understanding of what "coral" is, because it'll make things easier to write about.
6 Peonidin is only present in hybrid Anthuriums with some A. amnicola or A. formosum in their ancestry. The discovery of A. amnicola, in particular, did a lot for the world's Anthurium breeders, and it was still a relatively new and exciting thing in 1996 when the book was written. (You can hear the excitement in the text every time A. amnicola gets mentioned.)
Not only did A. amnicola introduce the possibility of purple spathes, but it's also been important in creating miniature Anthurium varieties, along with the white-to-lavender A. antioquense.
7 And then there are the yellow Anthuriums. You can't make yellow from orange, red, purple, and green. For our purposes, we're going to pretend that the yellows don't exist, because this is all complicated enough already, and anyway, they're probably explainable in the same sort of fashion that all the other colors are. The yellow pigment in most spadices comes from an unidentified (?) flavone (Pelargonidin, cyanidin, and peonidin are all anthocyanins, which are similar to flavones but not quite the same.), and there's a good chance that it's the same molecule in yellow spadices and spathes alike, but I haven't actually run into anything yet that explains yellow Anthuriums genetically, so if you ask me about them I'm probably going to just smile awkwardly, shrug, and try to change the subject.
Unless you're just asking whether or not they exist, in which case I will get very excited indeed. I've seen one in person, once, didn't buy it, have never seen another, and consequently have had a three-year-and-counting case of non-buyer's remorse that still gives me pangs of regret every time I remember.
It looked like this:
I'll also get very agitated if you ask me about dye-injected yellow Anthuriums. It's probably a good show, but stand back and bring a raincoat, as there will be spittle and gesticulations.
8 And there's a good chance that I have, because this was the trickiest part of the genetics and I spent a lot of time reading stuff about "recessive epistasis" over and over, hoping to understand it. I feel like I more or less do, but I still don't understand it quite well enough to explain it to people so there's a good chance that I'm not following something properly. Pretty much anything in this whole post could be wrong, but I'm at least trying pretty hard not to be wrong.
9 Which I know it wounds like "light red" should mean "pink," but I've seen it for myself, and "light red" is a different color. Damned if I know how to describe the difference to you, but #280 "Jujubee" is definitely light red, and #232 "Rhoda Badcek" is definitely pink, and they are definitely not the same color.
I think the distinction is mainly about the graininess of the color -- "light red" is a mix of full red speckles and white speckles, whereas pink is more uniform. That's at least how I distinguish them when I'm describing the colors to myself. The margins of light red spathes also tend to be fully red, whereas the margins of pink spathes may or may not be darker than the rest of the spathe, which you can also see in the above photos.
10 Which is also a thing that happens. 'White Gemini,' for example, is usually just white, as so:
But the last bloom I got from it had some streaks of pink at the base:
11 Nor Spathiphyllum or Schlumbergera, as far as that goes. Growing houseplants from seed is sometimes discussed a little bit, usually in the context of indoor/outdoor annual-type plants like coleus, but houseplants are a niche interest to begin with, and growing self-produced seeds of houseplants is even nichier.
12 pink: NOID pink, 'Pandola,' 'Joli'
red: NOID red, 'Gemini,' 'Red Hot'
red-violet: NOID red-violet, 'Krypton'
purple: NOID purple
orange: 'Florida,' 'Orange Hot' (if I'm feeling charitable)
white: 'White Gemini'
other: 'Peppermint Gemini' (blotchy red and white)
Though 'Joli' hasn't been around long enough to pollinate or have been pollinated; 'Florida' is possibly sterile (I've never seen it produce or accept pollen); I'm unclear about how different 'Gemini'/'White Gemini'/'Peppermint Gemini' actually are from one another; and there's a possibility that the NOID red-violet is also a 'Krypton.' (Neither one blooms very much, and I'm not sure they've ever been blooming simultaneously, so it's tough to make comparisons.) So it's possible that there are only really nine founding varieties.
13 Though good luck getting a green; none of the founding varieties are green, and I've seen no indication so far that any of them have genes for green spathes. If someone would like to buy me a 'Midori' so I can pursue this dream, feel free.
14 The leaves vary from one variety to the next a lot more than you'd think. Or possibly it's that thing where when you spend enough time looking closely at something, you eventually start making distinctions no one else would even see. Not sure which applies here, but I suppose we'll find out whenever I wind up talking about foliage.