I didn't believe this was a real thing, and I'm still very skeptical, but it is something that's claimed to happen!
Their stems & leaves have a siliceous epidermis, so if rubbing b/n 2 horsetails occurs, a forest fire may be produced
--The Bottled Ocean of Biology, by Nisarg Desai, 2017
Why don't other plants show same thing? Another plant, bamboo, is supposed to do the same thing:
How these fires originate is uncertain ... it has been suggested that a probable cause is the rubbing together of the silica-coated bamboo stems as they sway in the breeze.
--Burma, by R. Talbot Kelly (2004)
The same claim about bamboo is made in an Indian Government document on forest fire management.
What special thing about horsetail starts the fire? Apparently the claim is that it's the silica in the stems of the plant that is the important factor, and that horsetail and bamboo are both unusual in that regard. This does seem to be true:
... in general, ferns, gymnosperms and angiosperms accumulated less Si in their shoots than non-vascular plant species and horsetails.
--Phylogenetic Variation in the Silicon Composition of Plants
Bamboo isn't mentioned in that article, but the presence of silicon in bamboo is supported elsewhere:
One of the advantages of bamboo over wood is that bamboo contains some silicon in both inner surface (pith-ring) and outer surface (rind) of the bamboo culm.
--Calcium phosphate formation induced on silica in bamboo
As I say I remain very skeptical. There's no evidence shown that this has actually happened. Is it telling that all these claims come from the same general region (India and Burma)? My guess is that this is an old wives' tale that's been written down and uncritically accepted, but if someone has actually seen this happen I'd be happy to be proven wrong, because the idea of plants sparking out flames in the wind is a pretty cool one.
Edit to add a link claiming that bamboo-on-bamboo generates sparks due to the silicates:
The method consists of striking sparks out of the culm of Schizostachyum bamboo with flint, broken pottery, china or even another piece of bamboo. The sparks that occur by striking the Schizostachyum bamboo are presumably generated by the high silica content in this genus of bamboo.
--Bamboo Strike-A-Light, by Tom Lourens, Ash Kivilaakso and Ed Read (My emphasis)
Edit to add that the overall concept of plants starting their own fires has been seriously looked at; the author is skeptical:
Individual plant traits (such as leaf moisture content, retention of dead branches and foliage, oil rich foliage) are known to affect the flammability of plants but there is no evidence these characters evolved specifically to self-immolate, although some of these traits may have been secondarily modified to increase the propensity to burn. ... It is more parsimonious to conclude plants have evolved mechanisms to tolerate, but not promote, landscape fire.
--Have plants evolved to self-immolate?
Notably, in his list of factors potentially allowing self-immolation, he does not list silica, and horsetail isn't mentioned at all.