Is nepheline a feldspar?
No: nepheline, hardness 5.5 to 6, is a feldspathoid with too little silica to form feldspar. The eight-silica sum worked, and which test tells it from albite.
No. Nepheline, Na3K(Al4Si4O16) on the IMA list, is a feldspathoid: an aluminosilicate with too little silica to form feldspar. Worked, adding eight SiO2 gives exactly the composition of three albite and one orthoclase, 82.3 g of silica per 100 g. In the hand it resembles feldspar: hardness 5.5 to 6 against 6 to 6.5, and an overlapping weight.
In short
- The verdict. The Alex Strekeisen petrography atlas defines feldspathoids as aluminosilicates of sodium, potassium or calcium having too little silica to form feldspar, and names nepheline the most common. All 10 Handbook of Mineralogy feldspar sheets print a feldspar group line; the nepheline sheet prints none. The September 2026 IMA-CNMNC Master List lists species only: no line contains feldspar or feldspathoid.
- The eight-silica sum. Nepheline, Na3K(Al4Si4O16), carries 4 silicon for 4 alkalis; albite and orthoclase carry 3 silicon for each alkali. Add 8 SiO2 and the atoms come to Na3K Al4 Si12 O32, exactly 3 NaAlSi3O8 plus KAlSi3O8, both at 1064.98 per formula unit. Per 100 g of nepheline, that is 82.3 g of silica short of feldspar.
- Choose the test by the specimen. On a broken mass, cleavage: perfect in feldspar, poor in nepheline. Under a polarising microscope, the one-axis rule: nepheline uniaxial, every feldspar sheet biaxial. On the balance or the refractometer, nothing: the windows overlap or miss by 0.001. Hardness meets at 6.
- Where the one-axis rule fails. Kalsilite, which the kalsilite sheet describes as fine intergrowths with nepheline or potassic feldspar, is uniaxial negative too, with indices that overlap nepheline’s. Leucite is uniaxial positive but may be anomalously biaxial. The rule tells nepheline from feldspar, not from every feldspathoid.
- Why the question is asked. The nepheline sheet lists potassic feldspar and plagioclase first among its associates. USGS Mineral Commodity Summaries 2026 covers feldspar and nepheline syenite in one chapter and says in glass and pottery both function as a flux. A shared use and a shared rock are not a shared family.
| Test | Nepheline | Albite | Orthoclase | Microcline | Does it choose? | Wrong for (documented failure) |
|---|---|---|---|---|---|---|
| IMA formula; silicon per alkali | Na3K(Al4Si4O16); 1 | Na(AlSi3O8); 3 | K(AlSi3O8); 3 | K(AlSi3O8); 3 | Yes, with an analysis | Needs an analysis |
| SiO2 in the ideal formula | 41.13% | 68.74% | 64.76% | 64.76% | Yes, with an analysis | Anorthite, a feldspar, is 43.19%: silica alone does not define feldspar |
| Optic class: the one-axis rule | Uniaxial (−) | Biaxial (+) low, (−) high | Biaxial (−) | Biaxial (−) | Yes, with a polarising microscope | Kalsilite, uniaxial (−); anomalously biaxial leucite |
| Birefringence, worked | 0.003 to 0.004 | 0.008 to 0.011 | 0.004 to 0.005 | 0.007 to 0.010 | Not alone | Orthoclase meets nepheline at 0.004 |
| Refractive indices | 1.526 to 1.546 | 1.526 to 1.541 | 1.518 to 1.525 | 1.514 to 1.539 | No | Albite’s whole range lies inside nepheline’s; orthoclase misses it by 0.001 |
| Cleavage | Poor on {10-10}, {0001} | Perfect {001}, very good {010} | Perfect {001} and {010} | Perfect {001} and {010} | Yes, on a fresh break | Massive or altered material that shows no clean break |
| Hardness | 5.5 to 6 | 6 to 6.5 | 6 to 6.5 | 6 to 6.5 | No | Both reach 6 |
| 20.00 g piece in water | 2.55 to 2.66: 12.16 to 12.48 g | 2.60 to 2.65: 12.31 to 12.45 g | 2.55 to 2.63: 12.16 to 12.40 g | 2.54 to 2.57: 12.13 to 12.22 g | No | Every piece: all three windows touch nepheline’s |
Is nepheline a feldspar? What the sources say
Nepheline is not a feldspar: it is a feldspathoid, in the words of the Alex Strekeisen petrography atlas an aluminosilicate of sodium, potassium or calcium having too little silica to form feldspar. The atlas says feldspathoids take the place of feldspars in igneous rocks undersaturated with respect to silica, that they share framework structures of silica and alumina tetrahedra with the feldspars, but that the arrangement differs from species to species. It calls them not a true group, and names nepheline, leucite, sodalite and cancrinite.
What the reference works print. A script over the 10 Handbook of Mineralogy feldspar sheets in our cache, albite, oligoclase, andesine, labradorite, bytownite, anorthite, anorthoclase, microcline, orthoclase and sanidine, finds a feldspar group line on every one; the nepheline sheet prints no group line at all. The September 2026 IMA-CNMNC Master List has no line containing feldspar or feldspathoid: it lists species, such as nepheline, Na3K(Al4Si4O16), Rd, 2018 s.p., and albite, Na(AlSi3O8), G, 1815. The nepheline sheet and all ten feldspar sheets cite the same reference volume, Deer, Howie and Zussman’s Rock-forming minerals, volume 4, framework silicates, by script.
A framework, but not a zeolite either. Coombs et al., Canadian Mineralogist 35 (1997), the IMA zeolite report, say framework structures such as in feldspars, nepheline, sodalites and scapolites have channels too restricted for typical zeolitic behaviour and are not zeolites. The species page, nepheline, carries the sheet data and the Na3K check; they are not repeated here.
The eight-silica sum, worked
The eight-silica sum: nepheline plus eight SiO2 has exactly the composition of three albite and one orthoclase. Count the atoms. Na3K(Al4Si4O16) holds Na3 K1 Al4 Si4 O16; eight SiO2 add Si8 O16; the total, Na3 K1 Al4 Si12 O32, is 3 NaAlSi3O8 plus KAlSi3O8. On standard atomic weights both sides weigh 1064.98. Per 100 g of nepheline the missing silica is 82.3 g; for the sodium end, NaAlSiO4 plus 2 SiO2 equals NaAlSi3O8, it is 84.6 g.
What the sum means and does not mean. It is arithmetic on formulas: it shows how far short of feldspar nepheline is in silica, which is the atlas’s definition expressed as a number. It is not a recipe or a claim about any particular rock. The atlas says feldspathoids occur with feldspars but never with quartz; the nepheline sheet’s association list, potassic feldspar, plagioclase, sodic pyroxenes, sodic amphiboles, leucite, olivine, augite and diopside, has no quartz in it, by script.
Where silica alone fails. Silicon to aluminium is not the line: nepheline is 4:4, albite and orthoclase 3:1, but anorthite, Ca(Al2Si2O8), a feldspar, is 2:2 and 43.19% SiO2, close to nepheline’s 41.13%. What separates them is the structure the sheets and the atlas describe, not one ratio. The plagioclase series, albite to anorthite, is on is labradorite a feldspar.
Can you tell nepheline from feldspar by eye? Cleavage and lustre
On a fresh break, cleavage chooses: feldspar cleaves perfectly, nepheline poorly. The albite sheet prints cleavage perfect on {001}, very good on {010}; orthoclase and microcline perfect on {001} and {010}. The nepheline sheet prints cleavage poor on {10-10} and {0001}, fracture subconchoidal. A grain that breaks with curved surfaces and no flat, flashing planes is not behaving like feldspar. The flash test and its steps are on how to identify feldspar and are not repeated here.
Lustre and habit, as clues. The nepheline sheet gives vitreous to greasy, transparent to nearly opaque from inclusions or alteration, in stout six- or twelve-sided prisms, typically poorly formed. The albite, orthoclase and microcline sheets give vitreous, pearly on cleavages. Neither is a test on its own: the International Gem Society gives nepheline’s cleavage as indistinct, which fits the sheet, but a weathered feldspar can look greasy too.
Where the eye fails. Massive or granular nepheline shows no crystal shape, and an altered surface shows no cleavage. The sheet says nepheline is also granular, compact and massive; on such a piece, go to the microscope or to an analysis.
The one-axis rule: what the microscope shows
The one-axis rule: nepheline is uniaxial and every one of the 10 Handbook feldspar sheets is biaxial, so an optic figure separates them. The nepheline sheet gives uniaxial (−), ω = 1.529 to 1.546, ε = 1.526 to 1.542. The albite sheet gives biaxial (+) for the low form and (−) for the high; orthoclase and microcline biaxial (−). A script over the 10 sheets finds every optic class beginning with biaxial.
What the refractometer cannot do. Albite’s indices, 1.526 to 1.541, lie entirely inside nepheline’s 1.526 to 1.546, and microcline’s 1.514 to 1.539 overlap them; orthoclase’s 1.518 to 1.525 stop 0.001 below nepheline’s lowest, too close to decide anything at the bench. Birefringence, worked from the ranges, is 0.003 to 0.004 for nepheline, 0.004 to 0.005 for orthoclase, 0.007 to 0.010 for microcline and 0.008 to 0.011 for albite: low for all four, and orthoclase meets nepheline at 0.004. The gem society gives nepheline 0.004.
Where the rule fails. Kalsilite is uniaxial (−) on its sheet, with ω = 1.535 to 1.554, and occurs as fine intergrowths with nepheline or potassic feldspar. Leucite is uniaxial (+) but may be anomalously biaxial. Quartz is uniaxial (+), the opposite sign; it is on is quartz a mineral. Uniaxial says not feldspar; it does not say which feldspathoid.
Weight and hardness: what they cannot choose
The balance chooses nothing here. Worked as 20 − 20/D, a 20.00 g nepheline reads 12.16 to 12.48 g in water. Albite reads 12.31 to 12.45 g, orthoclase 12.16 to 12.40 g, microcline 12.13 to 12.22 g, and by script 7 of the 10 feldspar sheets overlap nepheline’s window. Quartz, at 12.45 g, sits inside it as well. The method is on how to measure specific gravity at home.
Hardness meets at 6. Nepheline is 5.5 to 6; the feldspars 6 to 6.5, sanidine and anorthoclase 6. A scratch that a feldspar resists and nepheline takes is possible near 6, but the ranges touch; see how to test mineral hardness properly.
The other feldspar questions. Labradorite’s composition test is on is labradorite a feldspar, sunstone’s refractometer bands on is sunstone a feldspar and amazonite’s index line on is amazonite a feldspar; none of them involves nepheline, and none is repeated here.
Who each test is wrong for
Each test short of an analysis fails on some specimen, so choose the one that fits the piece in hand. Cleavage is right for a fresh break and wrong for massive or altered material. The one-axis rule is right under a polarising microscope and silent against kalsilite. The balance and the scratch test are wrong for every nepheline-against-feldspar question, because their windows overlap; the refractometer is wrong against albite and microcline and leaves a gap of 0.001 against orthoclase. An analysis, read against the eight-silica sum, is decisive and rarely available.
The label. A white or grey grain in a syenite labelled feldspar may be nepheline, and the sheets give no hand test that settles it every time. Where the determination was not made, the honest label says so; how to annotate an old label without overwriting it is on how to label and catalogue a mineral collection.
On this site, the feldspathoid neighbours are sodalite and haüyne; the feldspar species pages are orthoclase and microcline; the classification questions are indexed at the collecting guides. Nothing on this site is offered for sale; the wanted list is the only commercial route.