diff --git a/lib/node_modules/@stdlib/math/base/special/exp10/lib/main.js b/lib/node_modules/@stdlib/math/base/special/exp10/lib/main.js
index a4c086eae198..8b6e17b692f2 100644
--- a/lib/node_modules/@stdlib/math/base/special/exp10/lib/main.js
+++ b/lib/node_modules/@stdlib/math/base/special/exp10/lib/main.js
@@ -38,7 +38,7 @@ var floor = require( '@stdlib/math/base/special/floor' );
 var ldexp = require( '@stdlib/math/base/special/ldexp' );
 var isnan = require( '@stdlib/math/base/assert/is-nan' );
 var MAXL10 = require( '@stdlib/constants/float64/max-base10-exponent' );
-var MINL10 = require( '@stdlib/constants/float64/min-base10-exponent' );
+var MINL10 = require( '@stdlib/constants/float64/min-base10-exponent-subnormal' );
 var PINF = require( '@stdlib/constants/float64/pinf' );
 var polyvalP = require( './polyval_p.js' );
 var polyvalQ = require( './polyval_q.js' );
diff --git a/lib/node_modules/@stdlib/math/base/special/exp10/manifest.json b/lib/node_modules/@stdlib/math/base/special/exp10/manifest.json
index 89e4c46efeae..02f1fe024445 100644
--- a/lib/node_modules/@stdlib/math/base/special/exp10/manifest.json
+++ b/lib/node_modules/@stdlib/math/base/special/exp10/manifest.json
@@ -42,7 +42,7 @@
         "@stdlib/constants/float64/pinf",
         "@stdlib/math/base/special/floor",
         "@stdlib/math/base/special/ldexp",
-        "@stdlib/constants/float64/min-base10-exponent",
+        "@stdlib/constants/float64/min-base10-exponent-subnormal",
         "@stdlib/constants/float64/nan"
       ]
     },
@@ -62,7 +62,7 @@
         "@stdlib/constants/float64/pinf",
         "@stdlib/math/base/special/floor",
         "@stdlib/math/base/special/ldexp",
-        "@stdlib/constants/float64/min-base10-exponent",
+        "@stdlib/constants/float64/min-base10-exponent-subnormal",
         "@stdlib/constants/float64/nan"
       ]
     },
@@ -82,7 +82,7 @@
         "@stdlib/constants/float64/pinf",
         "@stdlib/math/base/special/floor",
         "@stdlib/math/base/special/ldexp",
-        "@stdlib/constants/float64/min-base10-exponent",
+        "@stdlib/constants/float64/min-base10-exponent-subnormal",
         "@stdlib/constants/float64/nan"
       ]
     }
diff --git a/lib/node_modules/@stdlib/math/base/special/exp10/src/main.c b/lib/node_modules/@stdlib/math/base/special/exp10/src/main.c
index 42ff65360c46..e4743b0b3d7a 100644
--- a/lib/node_modules/@stdlib/math/base/special/exp10/src/main.c
+++ b/lib/node_modules/@stdlib/math/base/special/exp10/src/main.c
@@ -35,7 +35,7 @@
 #include "stdlib/math/base/special/ldexp.h"
 #include "stdlib/math/base/assert/is_nan.h"
 #include "stdlib/constants/float64/max_base10_exponent.h"
-#include "stdlib/constants/float64/min_base10_exponent.h"
+#include "stdlib/constants/float64/min_base10_exponent_subnormal.h"
 #include "stdlib/constants/float64/pinf.h"
 #include "stdlib/constants/float64/nan.h"
 #include <stdint.h>
@@ -129,7 +129,7 @@ double stdlib_base_exp10( const double x ) {
 	if ( x > STDLIB_CONSTANT_FLOAT64_MAX_BASE10_EXPONENT ) {
 		return STDLIB_CONSTANT_FLOAT64_PINF;
 	}
-	if ( x < STDLIB_CONSTANT_FLOAT64_MIN_BASE10_EXPONENT ) {
+	if ( x < STDLIB_CONSTANT_FLOAT64_MIN_BASE10_EXPONENT_SUBNORMAL ) {
 		return 0.0;
 	}
 	// Express 10^x = 10^g 2^n = 10^g 10^( n log10(2) ) = 10^( g + n log10(2) )
diff --git a/lib/node_modules/@stdlib/math/base/special/exp10/test/test.js b/lib/node_modules/@stdlib/math/base/special/exp10/test/test.js
index 7e6441bcf1f3..2645516ff2f3 100644
--- a/lib/node_modules/@stdlib/math/base/special/exp10/test/test.js
+++ b/lib/node_modules/@stdlib/math/base/special/exp10/test/test.js
@@ -25,6 +25,7 @@ var isnan = require( '@stdlib/math/base/assert/is-nan' );
 var PINF = require( '@stdlib/constants/float64/pinf' );
 var NINF = require( '@stdlib/constants/float64/ninf' );
 var isPositiveZero = require( '@stdlib/assert/is-positive-zero' );
+var isAlmostSameValue = require( '@stdlib/assert/is-almost-same-value' );
 var exp10 = require( './../lib' );
 
 
@@ -167,3 +168,23 @@ tape( 'the function returns `NaN` if provided `NaN`', function test( t ) {
 	t.strictEqual( isnan( val ), true, 'equals NaN' );
 	t.end();
 });
+
+tape( 'the function returns subnormal values for `x` in the subnormal range', function test( t ) {
+	var expected;
+	var x;
+	var v;
+	var i;
+
+	x = [ -308.5, -310.0, -320.0, -323.5 ];
+	expected = [
+		3.16227766016838e-309,
+		1e-310,
+		1e-320,
+		5e-324
+	];
+	for ( i = 0; i < x.length; i++ ) {
+		v = exp10( x[ i ] );
+		t.strictEqual( isAlmostSameValue( v, expected[ i ], 1 ), true, 'returns expected value' );
+	}
+	t.end();
+});
diff --git a/lib/node_modules/@stdlib/math/base/special/exp10/test/test.native.js b/lib/node_modules/@stdlib/math/base/special/exp10/test/test.native.js
index e3adf1d9f764..f643aa28240a 100644
--- a/lib/node_modules/@stdlib/math/base/special/exp10/test/test.native.js
+++ b/lib/node_modules/@stdlib/math/base/special/exp10/test/test.native.js
@@ -181,3 +181,23 @@ tape( 'the function returns `NaN` if provided `NaN`', opts, function test( t ) {
 	t.strictEqual( isnan( val ), true, 'equals NaN' );
 	t.end();
 });
+
+tape( 'the function returns subnormal values for `x` in the subnormal range', opts, function test( t ) {
+	var expected;
+	var x;
+	var v;
+	var i;
+
+	x = [ -308.5, -310.0, -320.0, -323.5 ];
+	expected = [
+		3.16227766016838e-309,
+		1e-310,
+		1e-320,
+		5e-324
+	];
+	for ( i = 0; i < x.length; i++ ) {
+		v = exp10( x[ i ] );
+		t.strictEqual( isAlmostSameValue( v, expected[ i ], 1 ), true, 'returns expected value' );
+	}
+	t.end();
+});
diff --git a/lib/node_modules/@stdlib/math/base/special/exp2/lib/main.js b/lib/node_modules/@stdlib/math/base/special/exp2/lib/main.js
index ed01a638747f..531dc55a568b 100644
--- a/lib/node_modules/@stdlib/math/base/special/exp2/lib/main.js
+++ b/lib/node_modules/@stdlib/math/base/special/exp2/lib/main.js
@@ -37,7 +37,7 @@
 // MODULES //
 
 var FLOAT64_MAX_BASE2_EXPONENT = require( '@stdlib/constants/float64/max-base2-exponent' ); // eslint-disable-line id-length
-var FLOAT64_MIN_BASE2_EXPONENT = require( '@stdlib/constants/float64/min-base2-exponent' ); // eslint-disable-line id-length
+var FLOAT64_MIN_BASE2_EXPONENT_SUBNORMAL = require( '@stdlib/constants/float64/min-base2-exponent-subnormal' ); // eslint-disable-line id-length
 var round = require( '@stdlib/math/base/special/round' );
 var ldexp = require( '@stdlib/math/base/special/ldexp' );
 var isnan = require( '@stdlib/math/base/assert/is-nan' );
@@ -104,7 +104,8 @@ function exp2( x ) {
 	if ( x > FLOAT64_MAX_BASE2_EXPONENT ) {
 		return PINF;
 	}
-	if ( x < FLOAT64_MIN_BASE2_EXPONENT ) {
+	// Note: values on the interval `(-1075, -1074)` round to the smallest subnormal...
+	if ( x <= FLOAT64_MIN_BASE2_EXPONENT_SUBNORMAL-1 ) {
 		return 0.0;
 	}
 	// Separate into integer and fractional parts...
diff --git a/lib/node_modules/@stdlib/math/base/special/exp2/manifest.json b/lib/node_modules/@stdlib/math/base/special/exp2/manifest.json
index a2adffecfdb4..3e41db2a664f 100644
--- a/lib/node_modules/@stdlib/math/base/special/exp2/manifest.json
+++ b/lib/node_modules/@stdlib/math/base/special/exp2/manifest.json
@@ -44,7 +44,7 @@
         "@stdlib/constants/float64/pinf",
         "@stdlib/math/base/special/round",
         "@stdlib/math/base/special/ldexp",
-        "@stdlib/constants/float64/min-base2-exponent"
+        "@stdlib/constants/float64/min-base2-exponent-subnormal"
       ]
     },
     {
@@ -64,7 +64,7 @@
         "@stdlib/constants/float64/pinf",
         "@stdlib/math/base/special/round",
         "@stdlib/math/base/special/ldexp",
-        "@stdlib/constants/float64/min-base2-exponent"
+        "@stdlib/constants/float64/min-base2-exponent-subnormal"
       ]
     },
     {
@@ -84,7 +84,7 @@
         "@stdlib/constants/float64/pinf",
         "@stdlib/math/base/special/round",
         "@stdlib/math/base/special/ldexp",
-        "@stdlib/constants/float64/min-base2-exponent"
+        "@stdlib/constants/float64/min-base2-exponent-subnormal"
       ]
     },
     {
@@ -104,7 +104,7 @@
         "@stdlib/constants/float64/pinf",
         "@stdlib/math/base/special/round",
         "@stdlib/math/base/special/ldexp",
-        "@stdlib/constants/float64/min-base2-exponent"
+        "@stdlib/constants/float64/min-base2-exponent-subnormal"
       ]
     }
   ]
diff --git a/lib/node_modules/@stdlib/math/base/special/exp2/src/main.c b/lib/node_modules/@stdlib/math/base/special/exp2/src/main.c
index a913467e70f6..4ba63c994df5 100644
--- a/lib/node_modules/@stdlib/math/base/special/exp2/src/main.c
+++ b/lib/node_modules/@stdlib/math/base/special/exp2/src/main.c
@@ -32,7 +32,7 @@
 
 #include "stdlib/math/base/special/exp2.h"
 #include "stdlib/constants/float64/max_base2_exponent.h"
-#include "stdlib/constants/float64/min_base2_exponent.h"
+#include "stdlib/constants/float64/min_base2_exponent_subnormal.h"
 #include "stdlib/math/base/special/round.h"
 #include "stdlib/math/base/special/ldexp.h"
 #include "stdlib/math/base/assert/is_nan.h"
@@ -128,7 +128,8 @@ double stdlib_base_exp2( const double x ) {
 	if ( x > STDLIB_CONSTANT_FLOAT64_MAX_BASE2_EXPONENT ) {
 		return STDLIB_CONSTANT_FLOAT64_PINF;
 	}
-	if ( x < STDLIB_CONSTANT_FLOAT64_MIN_BASE2_EXPONENT ) {
+	// Note: values on the interval `(-1075, -1074)` round to the smallest subnormal...
+	if ( x <= STDLIB_CONSTANT_FLOAT64_MIN_BASE2_EXPONENT_SUBNORMAL-1 ) {
 		return 0.0;
 	}
 
diff --git a/lib/node_modules/@stdlib/math/base/special/exp2/test/test.js b/lib/node_modules/@stdlib/math/base/special/exp2/test/test.js
index 8ebf5cba93a9..05fb27838ae5 100644
--- a/lib/node_modules/@stdlib/math/base/special/exp2/test/test.js
+++ b/lib/node_modules/@stdlib/math/base/special/exp2/test/test.js
@@ -134,6 +134,7 @@ tape( 'the function returns `+infinity` for very large `x`', function test( t )
 });
 
 tape( 'the function returns `0.0` for negative large `x`', function test( t ) {
+	t.strictEqual( exp2( -1075.0 ), 0.0, 'equals 0' );
 	t.strictEqual( exp2( -1100.0 ), 0.0, 'equals 0' );
 	t.strictEqual( exp2( -1200.0 ), 0.0, 'equals 0' );
 	t.strictEqual( exp2( -1300.0 ), 0.0, 'equals 0' );
@@ -168,3 +169,22 @@ tape( 'the function returns `NaN` if provided `NaN`', function test( t ) {
 	t.strictEqual( isnan( val ), true, 'equals NaN' );
 	t.end();
 });
+
+tape( 'the function returns subnormal values for `x` in the subnormal range', function test( t ) {
+	var expected;
+	var x;
+	var v;
+	var i;
+
+	x = [ -1023.5, -1050.0, -1074.0 ];
+	expected = [
+		7.866824069956793e-309,
+		8.289046e-317,
+		5e-324
+	];
+	for ( i = 0; i < x.length; i++ ) {
+		v = exp2( x[ i ] );
+		t.strictEqual( isAlmostSameValue( v, expected[ i ], 1 ), true, 'returns expected value' );
+	}
+	t.end();
+});
diff --git a/lib/node_modules/@stdlib/math/base/special/exp2/test/test.native.js b/lib/node_modules/@stdlib/math/base/special/exp2/test/test.native.js
index 0f928d20a0d6..d457065be763 100644
--- a/lib/node_modules/@stdlib/math/base/special/exp2/test/test.native.js
+++ b/lib/node_modules/@stdlib/math/base/special/exp2/test/test.native.js
@@ -143,6 +143,7 @@ tape( 'the function returns `+infinity` for very large `x`', opts, function test
 });
 
 tape( 'the function returns `0.0` for negative large `x`', opts, function test( t ) {
+	t.strictEqual( exp2( -1075.0 ), 0.0, 'equals 0' );
 	t.strictEqual( exp2( -1100.0 ), 0.0, 'equals 0' );
 	t.strictEqual( exp2( -1200.0 ), 0.0, 'equals 0' );
 	t.strictEqual( exp2( -1300.0 ), 0.0, 'equals 0' );
@@ -177,3 +178,22 @@ tape( 'the function returns `NaN` if provided `NaN`', opts, function test( t ) {
 	t.strictEqual( isnan( val ), true, 'equals NaN' );
 	t.end();
 });
+
+tape( 'the function returns subnormal values for `x` in the subnormal range', opts, function test( t ) {
+	var expected;
+	var x;
+	var v;
+	var i;
+
+	x = [ -1023.5, -1050.0, -1074.0 ];
+	expected = [
+		7.866824069956793e-309,
+		8.289046e-317,
+		5e-324
+	];
+	for ( i = 0; i < x.length; i++ ) {
+		v = exp2( x[ i ] );
+		t.strictEqual( isAlmostSameValue( v, expected[ i ], 1 ), true, 'returns expected value' );
+	}
+	t.end();
+});