为了使排序只适用于UNION中的第一个语句，你可以把它放在一个带有UNION ALL的子选择中(这两个在Oracle中都是必要的):

Select id,name,age FROM 
(    
 Select id,name,age
 From Student
 Where age < 15
 Order by name
)
UNION ALL
Select id,name,age
From Student
Where Name like "%a%"

或者(针对Nicholas Carey的评论)你可以保证顶部的SELECT是有序的，结果显示在下面的SELECT上面，就像这样:

Select id,name,age, 1 as rowOrder
From Student
Where age < 15
UNION
Select id,name,age, 2 as rowOrder
From Student
Where Name like "%a%"
Order by rowOrder, name

只写

Select id,name,age
From Student
Where age < 15
Union
Select id,name,age
From Student
Where Name like "%a%"
Order by name

顺序被应用到完整的结果集

可以用这个:

Select id,name,age
From Student
Where age < 15
Union ALL
SELECT * FROM (Select id,name,age
From Student
Where Name like "%a%")

Select id,name,age
from
(
   Select id,name,age
   From Student
   Where age < 15
  Union
   Select id,name,age
   From Student
   Where Name like "%a%"
) results
order by name

为了使排序只适用于UNION中的第一个语句，你可以把它放在一个带有UNION ALL的子选择中(这两个在Oracle中都是必要的):

Select id,name,age FROM 
(    
 Select id,name,age
 From Student
 Where age < 15
 Order by name
)
UNION ALL
Select id,name,age
From Student
Where Name like "%a%"

或者(针对Nicholas Carey的评论)你可以保证顶部的SELECT是有序的，结果显示在下面的SELECT上面，就像这样:

Select id,name,age, 1 as rowOrder
From Student
Where age < 15
UNION
Select id,name,age, 2 as rowOrder
From Student
Where Name like "%a%"
Order by rowOrder, name

其他两个答案都是正确的，但我认为值得注意的是，我陷入困境的地方是没有意识到你需要按别名排序，并确保两个选择的别名是相同的…所以

select 'foo'
union
select item as `foo`
from myTable
order by `foo`

请注意，我在第一个选择中使用单引号，而在其他选择中使用反引号。

这样就能得到你需要的排序。