In my study, I found that in volcanoes when the magma is going up it formed different types of rocks. There are basic, acidic and ultrabasic.

My question is why isn't there ultra acid igneous rocks formed?

  • 2
    $\begingroup$ The modern terminology uses the terms "mafic" and "felsic" instead of "basic" and "acidic", see Le Maitre et al. (2002), the bible of igneous rocks classification: doi.org/10.1017/CBO9780511535581 $\endgroup$ Aug 24, 2020 at 8:42
  • 2
    $\begingroup$ @Jean-MariePrival: so far you've had two people up vote your comment. It may be worth converting your comments into an answer. $\endgroup$
    – Fred
    Aug 24, 2020 at 20:18
  • $\begingroup$ @Fred Done, thanks for the suggestion. $\endgroup$ Aug 25, 2020 at 9:09

2 Answers 2


Warning: not an igneous petrologist here, I'm more interested in what happens to magmas once they reach the surface than all the processes they might have encountered along the way...

That being said, I think it has to do with the source of magmas, that is the mantle, which is itself mafic. Hence partial melting of the mantle generates mafic magmas. These magmas can then differentiate into more "evolved" (i.e., intermediate or felsic) ones. But there are limitations to these differentiation processes. With fractional crystallization you make mafic minerals, pumping Mg and Fe from the melt and leaving it silica-enriched, but you will never reach a 100 % silica melt (after all mafic minerals are still silicates, they need some Si in their structure). Same with crustal assimilation: the continental crust is only andesitic (intermediate) on average (Rudnick 1995), and even with assimilation of more felsic (i.e., granitic) portions you would never reach a 100 % silica melt. The starting material is too mafic.

On a more "philosophical" note, I would add that despite the absence of the term "ultrafelsic" in the literature, one could still consider some rocks as being ultrafelsic. If you look at the chemical classification of volcanic rocks (Le Maître et al. 2002), the four fields are separated like this (silica content in wt%):

  • Ultramafic: SiO$_2$ < 45
  • Mafic: 45 < SiO$_2$ < 52
  • Intermediate: 52 < SiO$_2$ < 63
  • Felsic: SiO$_2$ > 63

So everything above 63 wt% SiO$_2$ is just "felsic". But there are some rhyolites with more than 76 wt% SiO$_2$ (some of the Mono Domes for instance, see Table 2 in Bray et al. 2017)! So there is no distinction between a low-silica dacite and a high-silica rhyolite: they all are just "felsic". That doesn't mean they are not different, just that we never bothered making a fifth field. But those fields are just arbitrary, we could have split the silica spectrum into twelve fields, or just two.

  • 1
    $\begingroup$ You might want to point out the obsolete nature of the terms "acid" and "basic", simply because OP uses them. $\endgroup$
    – Spencer
    Aug 25, 2020 at 10:34

As mentioned in a comment to another answer, the terms "basic" and "acid" in relation to igneous rocks are outdated and stem from an incomplete understanding of rock chemistry in the past.

Better terms are "ultramafic", "mafic", "felsic", etc. As to your question - why aren't there "ultrafelsic" rocks? Let's put it into more quantitative numbers.

Felsic rocks (or acid, in the old terminology) are rocks with high SiO2 content, usually higher than 60% or 65%. Your hypothetical "ultrafelsic" rocks would then have maybe 85% or 90% SiO2.

The reason rocks like that form is because crystals with low SiO2 contents are removed from the melt, enriching the residual melt in SiO2 and push them to the "felsic" field.

At some point, the melt is sufficiently rich in SiO2 and quartz becomes a mineral that forms from the melt. Quartz is 100% SiO2, so it stops the melt from becoming richer in SiO2. If by some act of magic the melt reaches 90% SiO2, it simply crystallises more solid quartz, removing the excess SiO2 from the melt, and lowering it to more reasonable ~75%.


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.