desertnance, are those your own numbers or were you just doing punnet squares? because either way i'm not really understanding your numbers over here. Either you misunderstand the punnet squares (with which I can help) or, if those are actual hard numbers observed from crosses that tells us that this characteristic does not inherit with cut and dry Mendellian ratios.
You must remember with punnet squares that you must make every possible pair. so if you cross parent A1A2 to parent a3a4, you have to have all possible combinations of A1 with a3 and a4. Then you need all possible combinations of A2 with a3 and a4. In your example above, you had AA x aa, which, using this little pairing method i just described will give all 4 possible genotypes of offspring be Aa heterozygotes.
Okay, just to lend myself a bit of credibility, I'm 5/8ths of the way through my Biology B.A. and just finished Organic Evolution course and Elements of Genetics courses.
If you were to cross an aa (autoflower, recessive/mutant) x AA (photoperiod, dominant/wild-type), no matter what you could do, all possible genotypes in the f1 are heterozygotes. No homozygous dominant, no homozygous recessive.
So either you fudged the punnet squares, or (if those are actual observed numbers) you have to reject the null hypothesis (hypothesis being that plants will show mendellian inheritance) because there is something else behind the curtains that we're not seeing.
Usually when characteristics aren't inherited in a cut and dry Mendellian fashion it is because of multiple genes interacting to produce a phenotype. Here are some of the possible cases:
-Mendellian- you observe 3 dominant for every 1 recessive. this would be characteristic of Mendellian inheritance of one gene for one trait
-codominance- instead of two phenotypes (auto and non-auto) there is now a 3rd phenotype that is a blend of the two. this is usually a 1:2:1 ratio and shows that dominant and recessive combine to make a new phenotype. (this is what i have personally observed in autoflower photo crosses)
-epistasis- when the expression of one gene pair masks or modifies another. ex would be coat color in mice. at locus A, wild type color is agouti (AA, Aa), which is dominant to black (aa). But,when recessive at locus B, no matter the genotype at A, the mouse loses all pigmentation and is albino.
So WIZ to get to your question - I have never seen autoflowers in the f1 generation, and I really have no idea how to explain you seeing it without some numbers like what percentage show it, etc.
Right off the top of my head I thought that You could have used a heterozygous photoperiod plant to cross with the autoflowers. One of the theories behind autoflowers is that it's a weird gene mutation that showed up when people were heavily breeding indicas in the 80s for the first to show sex and the fastest to flower.
So, possibly you bred your autoflower to a short-night indica? Many of our modern indoor indicas, I have heard, can sometimes be heterozygous for the autoflower gene - meaning that it causes them to flower with less night hours than their brethren. If this were the case, and if the autoflower gene inherited according to Mendellian genetics, you would see about 50% autoflower in the f1. Is this what you saw? About 50% auto?
However, if you saw only 1 out of 100 autoflower or something along those lines, then I have no idea for you. However, if you saw a statistically significant amount (5% or more auto in f1) then, based on the numbers you give me we can find the most probable method of inheritance.
