Is howlite turquoise?
No. Howlite is a white calcium borosilicate, hardness 3.5; turquoise a copper phosphate, 5 to 6. GIA’s turquoise SG range holds howlite’s; weight cannot decide.
No. Howlite is a calcium borosilicate, Ca2SiB5O9(OH)5, white and hardness 3.5; turquoise is a copper aluminium phosphate, hardness 5 to 6. Blue howlite is dyed howlite. Weight cannot choose: GIA’s turquoise range, SG 2.40 to 2.90, contains all of howlite’s 2.53 to 2.59. Refractive index can: every howlite index is 1.605 or less, and GIA’s turquoise begins at 1.610.
In short
- Two species, two classes. The September 2026 IMA-CNMNC Master List lists howlite as Ca2SiB5O9(OH)5 and turquoise as CuAl6(PO4)4(OH)8(H2O)4. Worked on standard atomic weights, howlite is 13.81% boron with no copper, turquoise 7.81% copper with no boron. The howlite sheet gives white as its only colour, so blue howlite has been dyed.
- The 1.61 line. The howlite sheet and the International Gem Society panel give α 1.583 to 1.586, β 1.596 to 1.598, γ 1.605. The GIA turquoise page gives RI 1.610 to 1.650, and the turquoise sheet α 1.61, γ 1.65. A reading under 1.61 is outside turquoise on both; the gap is only 0.005, so this is a refractometer decision, not a hand test.
- The weight trap, documented. GIA gives turquoise SG 2.76 (+0.14, −0.36): 2.40 to 2.90, or 11.67 to 13.10 g in water for a 20.00 g piece. Howlite reads 12.09 to 12.28 g on its sheet and 11.84 to 12.25 g on the gem society’s range: inside GIA’s window both times. The turquoise sheet’s own 2.86 (13.01 g) is 0.73 g away, but it adds less if massive.
- What this page does not repeat. The scratch test is set out on turquoise and how to tell real turquoise from fake; the dye-location check on how to tell if a crystal is dyed; the 2.63 line on is chrysocolla the same as turquoise. Those numbers agree with this page’s: howlite 3.5, turquoise 5 to 6.
- Wrong for whom. “Turquoise” is wrong for a white stone with grey veins and for any stone reading under 1.61. “Howlite” is wrong for a stone reading 1.61 or more. “Dyed howlite” is the correct label for blue howlite, and the gem society calls it a most convincing simulant.
| Criterion | Howlite (sheet) | Howlite (IGS) | Turquoise (sheet) | Turquoise (GIA) | Does it choose? |
|---|---|---|---|---|---|
| Formula | Ca2B5SiO9(OH)5 (IMA: Ca2SiB5O9(OH)5) | Ca2B5SiO9(OH)5 | CuAl6(PO4)4(OH)8·4H2O (IMA: CuAl6(PO4)4(OH)8(H2O)4) | Hydrated copper and aluminium phosphate (description page) | Yes, by laboratory analysis |
| Chemical class | Borosilicate | — | Phosphate | Phosphate | Yes, by laboratory analysis |
| Boron / copper by mass | 13.81% B, no Cu | — | 7.81% Cu, no B | — | Yes, by laboratory analysis |
| Natural colour | White | White; black threadlike impurities | Sky-blue, bluish green, apple-green, greenish grey | Blue to green | Yes: blue howlite is dyed |
| Refractive index | 1.583–1.605 | 1.583–1.605 | 1.61–1.65 | 1.610–1.650 | Yes: the 1.61 line, a 0.005 gap |
| Hardness | 3.5 | 3–3.5 | 5–6 | 5–6 | Yes; method on the linked pages |
| Density / SG | 2.53–2.59 | 2.45–2.58 | 2.86, less if massive | 2.40–2.90 | No: GIA’s range contains howlite’s |
| 20.00 g piece in water | 12.09–12.28 g | 11.84–12.25 g | 13.01 g | 11.67–13.10 g | No, against GIA’s range |
| UV | Nothing recorded | Brownish yellow SW; some California deep orange LW | Nothing recorded | Not stated | No |
| Crystals | Tabular, to 1 cm; most commonly nodular | Tiny; nodular masses | Rare, to 3 mm | Aggregate of microscopic crystals | No: both usually massive |
| Setting | Borate deposits | Arid regions or borate deposits | Porphyry copper alteration; volcanic veins; phosphate-rich sediments | Dry regions, copper-rich groundwater | Yes, if the locality is known |
Is howlite the same as turquoise? The criteria
Howlite is not turquoise: one is a calcium borosilicate and the other a copper aluminium phosphate, and they share no essential element except oxygen and hydrogen. The Handbook howlite sheet and the turquoise sheet give Ca2B5SiO9(OH)5 and CuAl6(PO4)4(OH)8·4H2O; the September 2026 IMA list writes Ca2SiB5O9(OH)5 and CuAl6(PO4)4(OH)8(H2O)4. Worked on standard atomic weights, howlite is 13.81% boron, turquoise 7.81% copper.
Four criteria, in the order that settles it. Natural colour: howlite is white, so a blue one is dyed. Refractive index: the 1.61 line. Hardness: 3.5 against 5 to 6, tested as the linked pages describe. Weight: not decisive against GIA’s turquoise range. Chemistry decides everything, but only in a laboratory.
Why the question exists. The International Gem Society says howlite is frequently dyed blue to resemble turquoise and makes a most convincing simulant, and that its black, threadlike impurities resemble the veining in turquoise. GIA’s turquoise description calls that veining matrix, remnants of host rock, and the fine-lined form spiderweb turquoise. A dyed howlite nodule can therefore copy both the colour and the web.
Howlite vs turquoise by refractive index: the 1.61 line
The 1.61 line: on the sources read here, every howlite refractive index is 1.605 or less and every turquoise figure is 1.610 or more, so a reading under 1.61 is outside turquoise. Howlite is biaxial (−) with α 1.583 to 1.586, β 1.596 to 1.598 and γ 1.605 on both its Handbook sheet and the gem society panel. GIA gives turquoise 1.610 to 1.650 and the Handbook α 1.61, β 1.62, γ 1.65.
Where the line is thin. The gap between howlite’s highest figure and GIA’s lowest is 0.005. Both materials are usually aggregates, so a reading on a polished nodule is a spot reading, not the full range of a crystal. GIA names turquoise among the gems most commonly encountered impregnated with wax or plastic, and this page has no source for how that shifts a reading, so it does not claim the line survives treatment. It is a laboratory decision; it is not offered as a home test.
Can weight tell howlite from turquoise? The weight trap
Not against GIA’s turquoise. The GIA turquoise facts give SG 2.76 (+0.14, −0.36), 2.40 to 2.90. Worked as 20 − 20/SG, a 20.00 g piece reads 11.67 to 13.10 g in water as turquoise. Howlite reads 12.09 to 12.28 g at the sheet’s 2.53 to 2.59, and 11.84 to 12.25 g at the gem society’s 2.45 to 2.58. Documented failure: every howlite reading falls inside GIA’s turquoise window.
Why the sheets look better than they are. Against the turquoise sheet’s single 2.86, 13.01 g, howlite is 0.73 g lighter, which is easy to see on a 0.01 g scale. But the sheet itself says less if massive, and GIA’s lower bound, 2.40, is below howlite’s. The live turquoise page reached the same conclusion from the sheets; GIA’s range quantifies it. Weighing in water is set out on how to measure specific gravity at home.
Who each answer is wrong for
“Turquoise” is wrong for a white or grey-veined white stone, which is the howlite sheet’s colour; for any stone reading under 1.61 on a refractometer; and for a stone with blue only along its cracks and veins, which is a dye pattern described on how to tell if a crystal is dyed.
“Howlite” is wrong for a stone reading 1.61 or more, and for a piece with a documented copper-district locality rather than a borate deposit. It is incomplete for any blue stone: the accurate label is dyed howlite. Separating dyed howlite from dyed magnesite is on is magnesite the same as howlite.
Neither answer is complete for wax- or polymer-impregnated turquoise, which GIA covers as a treatment; the identification guides are in the collecting guides, and the mineral itself is on howlite. Nothing on this site is offered for sale; the wanted list is the only commercial route.