I think there are several things going on. What I was fascinated by in my research was how six photos of a tea rose could look so completely different that you would swear they were of different roses. It was true of many tea roses. And this, I think, is why.
One can always go back to Mutablilis and view a bud opening yellow, turning an orangish color, then pinkish, then reddish. There are two color processes working at once there: a fading yellow and a developing red. And all of the colors in between can be explained by just where each process is at the time. Of course everything that might affect the metabolism of the rose could affect the speed of these processes, perhaps a little differently: light, temperature, moisture, environmental stresses, and so on. My guess is that Mutabilis was used quite a bit in the breeding of tea roses and that many tea roses inherited and express both color processes. But evidently other heritable traits moderate to what extent the processes are expressed, so that the pigments produced by the processes are the same but perhaps how the pigments are distributed in the petals is different. And the amount of pigment produced under similar conditions varies from cultivar to cultivar.
Consider Madame Lombard
. In this photo
a newer blossom is apricot and an older one is a darker pink. One can find photos of Madame Lombard that depict a pale rose, almost white
or a dark rosy shade with orange-tinged central petals and darker outer petals.
. This can be explained by the progress of those same two color processes found in Mutabilis. What we see in Madame Lombard is happening also in Smoky. But Smoky starts with a third color process, evidently, one that very reliably and uniformly produces rose or magenta pigment as a baseline color throughout the petal rather than the white baseline in the case of Mme Lombard.
So all of this color variation in (some number of) tea roses can be explained in terms of two known petal coloring processes traced back to Mutabilis. Those two processes work with a third "uniform pink" process in Smoky. All kinds of environmental factors could, hypothetically, affect the speed of each process. Every rose on a given plant that experiences precisely the same amount of light and that is formed on the same day is likely to resemble every other rose. But as environmental factors change, and the processes run differently, the color of the rose will be different. Maybe imperceptibly, maybe much. Most people claim that colors are richer in their fall roses. Colors that derive from the reddening color process of Mutabilis and other china roses might develop more slowly in cooler weather. So cooler weather might favor that coveted orange coloration on Smoky's central petals.
Because there exist a number of hybrid tea roses such as Ophelia and Peace that sport, causing quasi-permanent quantitative changes in the magnitude of expression of these color processes, it must be true that a sport of Smoky could decrease (or increase) the expression of yellow pigment leading to less (or more) orange and the impression that it's not the real Smoky. But another possibility is that environmental factors don't favor the creation of yellow pigment because that process can be really tweaky in a lot of roses that have it. Perhaps some change in the plant's nutrition, water rations, light, or thermal environment would bring that yellow pigment back.
At this point I need more information. Experiments are indicated. Mme Lombard is now on the ARE to-buy list.