diff --git a/src/FillArrays.jl b/src/FillArrays.jl
index bf090490..f07d2a56 100644
--- a/src/FillArrays.jl
+++ b/src/FillArrays.jl
@@ -262,6 +262,7 @@ for (Typ, funcs, func) in ((:Zeros, :zeros, :zero), (:Ones, :ones, :one))
         @inline $Typ{T,N}(A::AbstractArray{V,N}) where{T,V,N} = $Typ{T,N}(size(A))
         @inline $Typ{T}(A::AbstractArray) where{T} = $Typ{T}(size(A))
         @inline $Typ(A::AbstractArray) = $Typ{eltype(A)}(A)
+        @inline $Typ(::Type{T}, m...) where T = $Typ{T}(m...)
 
         @inline axes(Z::$Typ) = Z.axes
         @inline size(Z::$Typ) = length.(Z.axes)
diff --git a/src/fillalgebra.jl b/src/fillalgebra.jl
index 2dec1b61..800e803a 100644
--- a/src/fillalgebra.jl
+++ b/src/fillalgebra.jl
@@ -86,7 +86,6 @@ end
 *(a::ZerosMatrix, b::AbstractMatrix) = mult_zeros(a, b)
 *(a::AbstractMatrix, b::ZerosVector) = mult_zeros(a, b)
 *(a::AbstractMatrix, b::ZerosMatrix) = mult_zeros(a, b)
-*(a::ZerosVector, b::AbstractVector) = mult_zeros(a, b)
 *(a::ZerosMatrix, b::AbstractVector) = mult_zeros(a, b)
 *(a::AbstractVector, b::ZerosMatrix) = mult_zeros(a, b)
 
diff --git a/test/runtests.jl b/test/runtests.jl
index 81ba61c5..35b647e2 100644
--- a/test/runtests.jl
+++ b/test/runtests.jl
@@ -20,6 +20,7 @@ include("infinitearrays.jl")
 
             for T in (Int, Float64)
                 Z = $Typ{T}(5)
+                @test $Typ(T, 5) ≡ Z
                 @test eltype(Z) == T
                 @test Array(Z) == $funcs(T,5)
                 @test Array{T}(Z) == $funcs(T,5)
@@ -34,6 +35,7 @@ include("infinitearrays.jl")
                 @test $Typ(2ones(T,5)) == Z
 
                 Z = $Typ{T}(5, 5)
+                @test $Typ(T, 5, 5) ≡ Z
                 @test eltype(Z) == T
                 @test Array(Z) == $funcs(T,5,5)
                 @test Array{T}(Z) == $funcs(T,5,5)
@@ -508,9 +510,9 @@ end
     @test_throws MethodError [1,2,3]*Zeros(3) # Not defined for [1,2,3]*[0,0,0] either
 
     @testset "Check multiplication by Adjoint vectors works as expected." begin
-        @test randn(4, 3)' * Zeros(4) === Zeros(3)
-        @test randn(4)' * Zeros(4) === zero(Float64)
-        @test [1, 2, 3]' * Zeros{Int}(3) === zero(Int)
+        @test randn(4, 3)' * Zeros(4) ≡ Zeros(3)
+        @test randn(4)' * Zeros(4) ≡ transpose(randn(4)) * Zeros(4) ≡ zero(Float64)
+        @test [1, 2, 3]' * Zeros{Int}(3) ≡ zero(Int)
         @test [SVector(1,2)', SVector(2,3)', SVector(3,4)']' * Zeros{Int}(3) === SVector(0,0)
         @test_throws DimensionMismatch randn(4)' * Zeros(3)
         @test Zeros(5)' * randn(5,3) ≡ Zeros(5)'*Zeros(5,3) ≡ Zeros(5)'*Ones(5,3) ≡ Zeros(3)'